8 Data encryption Meno Abels (see projects)
9 Communication encryption Meno Abels
10 Version 1.35 Kern (see below)
11 ========================================================
13 1.35 Items to do for release:
14 - Fix silly restriction requiring Include { Options { xxx } } to be
16 - Improve error message if old/new FileSet syntax mixed.
17 - Restore c: with a prefix into /prefix/c/ to prevent c: and d:
18 files with the same name from overwritting each other.
19 - Add new DCR calling sequences everywhere in SD. This will permit
20 simultaneous use of multiple devices by a single job.
23 - Document a get out of jail procedure if everything breaks if
24 you lost/broke the Catalog -- do the same for "I know my
25 file is there how do I get it back?".
27 - New IP address specification is used as follows:
28 [sdaddresses|diraddresses|fdaddresses] = { [[ip|ipv4|ipv6] = {
29 [[addr|port] = [^ ]+[\n;]+] }] }
31 so it could look for example like this:
32 SDaddresses = { ip = {
33 addr = 1.2.3.4; port = 1205; }
35 addr = 1.2.3.4; port = http; }
51 addr = bluedot.thun.net
54 as a consequence, you can now specify multiple IP addresses and
55 ports to be used. In the case of a server, it will listen on
56 all those that you specify. In the case of connecting to the server,
57 Bacula will attempt connecting to one at a time until it succeeds.
58 And, in a few other special cases, Bacula will use only the first
61 The default port numbers are still the same and the services and hosts
62 are also resolved by name. So now you could use the real names for the
65 An ip section will allow resolution to either an ipv4 or an ipv6 address.
66 An ipv4 section forces the resolution to be only ipv4, and an ipv6 section
67 forces the resolution to be only ipv6.
71 - Look at patches/bacula_db.b2z postgresql that loops during restore.
73 - Add delete JobId to regression.
74 - Add bscan to four-concurrent-jobs regression.
75 - Add IPv6 to regression
76 - Perhaps add read/write programs and/or plugins to FileSets.
79 Documentation to do: (any release a little bit at a time)
80 - Alternative to static linking "ldd prog" save all binaries listed,
81 restore them and point LD_LIBRARY_PATH to them.
82 - Document add "</dev/null >/dev/null 2>&1" to the bacula-fd command line
83 - Document query file format.
84 - Add more documentation for bsr files.
85 - Document problems with Verify and pruning.
86 - Document how to use multiple databases.
87 - VXA drives have a "cleaning required"
88 indicator, but Exabyte recommends preventive cleaning after every 75
91 In this context, it should be noted that Exabyte has a command-line
92 vxatool utility available for free download. (The current version is
93 vxatool-3.72.) It can get diagnostic info, read, write and erase tapes,
94 test the drive, unload tapes, change drive settings, flash new firmware,
96 Of particular interest in this context is that vxatool <device> -i will
97 report, among other details, the time since last cleaning in tape motion
98 minutes. This information can be retrieved (and settings changed, for
99 that matter) through the generic-SCSI device even when Bacula has the
100 regular tape device locked. (Needless to say, I don't recommend
101 changing tape settings while a job is running.)
102 - Lookup HP cleaning recommendations.
103 - Lookup HP tape replacement recommendations (see trouble shooting autochanger)
104 - Add more obvious documentation of @ for conf files.
105 - Document doing table repair
108 Testing to do: (painful)
111 For 1.37 Testing/Documentation:
112 - Add "Allow multiple connections" in Catalog resource to open a new
113 database connection for each job.
114 - Allow Simultaneous Priorities = yes => run up to Max concurrent jobs even
115 with multiple priorities.
116 - Fix find_device in stored/dircmd.c:462 (see code)
117 - Add db check test to regression. Test each function like delete,
119 - If you use restore replace=never, the directory attributes for
120 non-existent directories will not be restored properly.
125 mt status and tapeinfo output are attached. Default blocksize was
126 originally 512 bytes. I ran the following test on my machine:
128 - set defblksize to 512
129 - started bacula, labelled tape, ran job
130 - job completes with the following in log
131 SD termination status: OK
132 Termination: Backup OK
133 - restore attempts with blocksize set to 0 or 512 result in the familiar
134 "Volume data error! Wanted block-id BB02, got . Buffer discarded."
137 # tapeinfo -f /dev/sg0
138 Product Type: Tape Drive
140 Product ID: 'TSL-11000 '
142 Attached Changer: Yes
143 SerialNumber: '0010000443'
150 Medium Type: Not Loaded
155 DataDeCompEnabled: yes
161 # mt -f /dev/nst0 status
163 File number=0, block number=132, partition=0.
164 Tape block size 512 bytes. Density code 0x26 (DDS-4 or QIC-4GB).
165 Soft error count since last status=0
166 General status bits on (1010000):
172 - see lzma401.zip in others directory for new compression
174 - Minimal autochanger handling in Bacula and in btape.
175 - Look into how tar does not save sockets and the possiblity of
176 not saving them in Bacula (Martin Simmons reported this).
177 The next two lines will show them.
178 localmounts=`awk '/ext/ { print $2 }' /proc/mounts` # or whatever
179 find $localmounts -xdev -type s -ls
180 - Fix restore jobs so that multiple jobs can run if they
181 are not using the same tape(s).
182 - Allow the user to select JobType for manual pruning/purging.
183 - Look at adding Client run command that will use the
184 port opened by the client.
185 - bscan does not put first of two volumes back with all info in
187 - Implement the FreeBSD nodump flag in chflags.
188 - Figure out how to make named console messages go only to that
189 console and to the non-restricted console (new console class?).
190 - Make restricted console prompt for password if *ask* is set or
191 perhaps if password is undefined.
192 - Implement "from ISO-date/time every x hours/days/weeks/months" in
194 ==== from Marc Schoechlin
195 - the help-command should be more verbose
196 (it should explain the paramters of the different
198 -> it´s time-comsuming to consult the manual anytime
199 you need a special parameter
200 -> maybe it´s more easy to maintain this, if the
201 descriptions of that commands are outsourced to
203 - the cd-command should allow complete paths
204 i.e. cd /foo/bar/foo/bar
205 -> if a customer mails me the path to a certain file,
206 it´s faster to enter the specified directory
207 - if the password is not configured in bconsole.conf
208 you should be asked for it.
209 -> sometimes you like to do restore on a customer-machine
210 which shouldn´t know the password for bacula.
211 -> adding the password to the file favours admins
212 to forget to remove the password after usage
214 the protection of that file is less important
215 - long-listed-output of commands should be scrollable
216 like the unix more/less-command does
217 -> if someone runs 200 and more machines, the lists could
218 be a little long and complex
219 - command-output should be shown column by column
220 to reduce scrolling and to increase clarity
222 - lsmark should list the selected files with full
224 - wildcards for selecting and file and directories would be nice
225 - any actions should be interuptable with STRG+C
226 - command-expansion would be pretty cool
228 - Compaction of Disk space by "migrating" Volumes that have pruned
229 Jobs (what criteria? size, #jobs, time).
230 - Add prune all command
231 - Document fact that purge can destroy a part of a restore by purging
232 one volume while others remain valid -- perhaps mark Jobs.
233 - Add multiple-media-types.txt
234 - look at mxt-changer.html
235 - Make ? do a help command (no return needed).
236 - Implement restore directory.
237 - Add All Local Partitions = yes to new style saves.
238 - Document streams and how to implement them.
239 - Possibly implement "Ensure Full Backup = yes" looks for a failed full backup
240 and upgrades the current backup if one exists.
241 - Check that barcode reading and update slots scan works.
242 - Try not to re-backup a file if a new hard link is added.
243 - Add feature to backup hard links only, but not the data.
244 - Add "All Local = yes" option to save to include all local partitions.
245 - Fix stream handling to be simpler.
246 - Add Priority and Bootstrap to Run a Job.
247 - Eliminate Restore "Run Restore Job" prompt by allowing new "run command
249 - Remove View FileSet button from Run a Job dialog.
250 - Don't count higher level non-saved directories in Files expected
252 - Handle prompt for restore job at end of Restore command.
253 - Add display of total selected files to Restore window.
254 - Add tree pane to left of window.
255 - Add progress meter.
256 - Polling does not work for restore. It tries a number of times,
257 gives up, and crashes the SD.
258 - Lock jcr_chain when doing attach/detach in acquire.c
259 - Add assert in free_jcr if attach/detach chain active.
260 - Max wait time or max run time causes seg fault -- see runtime-bug.txt
261 - Document writing to a CD/DVD with Bacula.
262 - Add check for tape alerts.
263 - Add a "base" package to the window installer for pthreadsVCE.dll
264 which is needed by all packages.
265 - Add message to user to check for fixed block size when the forward
266 space test fails in btape.
267 - When unmarking a directory check if all files below are unmarked and
268 then remove the + flag -- in the restore tree.
269 - Possibly implement: Action = Unmount Device="TapeDrive1" in Admin jobs.
270 - Setup lrrd graphs: (http://www.linpro.no/projects/lrrd/) Mike Acar.
271 - Revisit the question of multiple Volumes (disk) on a single device.
272 - Finish SIGHUP work.
273 - Check that all change in wait status in the SD are
274 signaled to the Director.
275 - Add a block copy option to bcopy.
276 - Investigate adding Mac Resource Forks.
277 - Finish work on Gnome restore GUI.
278 - Fix "llist jobid=xx" where no fileset or client exists.
279 - Check pruning of restore jobs.
281 it seems to be complaining about 12:00pm which should be a valid 12
282 hour time. I changed the time to 11:59am and everything works fine.
283 Also 12:00am works fine. 0:00pm also works (which I don't think
284 should). None of the values 12:00pm - 12:59pm work for that matter.
285 - Add level to estimate command.
286 - For each job type (Admin, Restore, ...) require only the really necessary
288 - Fix option 2 of restore -- list where file is backed up -- require Client,
289 then list last 20 backups.
290 - Add all pools in Dir conf to DB also update them to catch changed
291 LabelFormats and such.
292 - Update volumes FromPool=xxx does all volumes.
293 - Pass Director resource name as an option to the Console.
294 - Add a "batch" mode to the Console (no unsolicited queries, ...).
295 - Add code to check for tape alerts -- tapeinfo.
296 - Make sure list of Volumes needed is in correct order for restore.
298 - Remove paths (and files that reference them) that have no trailing slash
299 in dbcheck -- or add a trailing slash.
300 - Remove Filenames (and files that reference them) that have a trailing
301 slash in dbcheck -- or remove the trailing slash.
302 - Remove orphaned paths/filenames by copying them to a new table with a
303 reference count, then mark all referenced files/paths and remove unreferenced
305 - Add a .list all files in the restore tree (probably also a list all files)
306 Do both a long and short form.
307 - Allow browsing the catalog to see all versions of a file (with
308 stat data on each file).
309 - Restore attributes of directory if replace=never set but directory
311 - Allow "delete job jobid=xxx,yyy,aaa-bbb" i.e. list + ranges.
312 - Use SHA1 on authentication if possible.
313 - See comtest-xxx.zip for Windows code to talk to USB.
314 - Make btape accept Device Names in addition to Archive names.
315 - Add Events and Perl scripting.
316 - Add John's appended files:
317 Appended = { /files/server/logs/http/*log }
318 and such files would be treated as follows.On a FULL backup, they would
319 be backed up like any other file.On an INCREMENTAL backup, where a
320 previous INCREMENTAL or FULL was already in thecatalogue and the length
321 of the file wasgreater than the length of the last backup, only thedata
322 added since the last backup will be dumped.On an INCREMENTAL backup, if
323 the length of the file is less than thelength of the file with the same
324 name last backed up, the completefile is dumped.On Windows systems, with
325 creation date of files, we can be evensmarter about this and not count
326 entirely upon the length.On a restore, the full and all incrementals
327 since it will beapplied in sequence to restore the file.
328 - Add a regression test for dbcheck.
329 - Add disk seeking on restore. - Allow
330 for optional cancelling of SD and FD in case DIR
331 gets a fatal error. Requested by Jesse Guardiani <jesse@wingnet.net>
332 - Bizarre message: Error: Could not open WriteBootstrap file:
333 - Build console in client only build.
334 - Add "limit=n" for "list jobs"
335 - Check new HAVE_WIN32 open bits.
336 - Check if the tape has moved before writing.
337 - Handling removable disks -- see below:
338 - Multiple drive autochanger support -- see below.
339 - Keep track of tape use time, and report when cleaning is necessary.
340 - Fix FreeBSD mt_count problem.
341 - Add FromClient and ToClient keywords on restore command (or
342 BackupClient RestoreClient).
343 - Automatic "update slots" on user configuration directive when a
345 - Implement a JobSet, which groups any number of jobs. If the
346 JobSet is started, all the jobs are started together.
347 Allow Pool, Level, and Schedule overrides.
348 - Enhance cancel to timeout BSOCK packets after a specific delay.
349 - When I restore to Windows the Created, Accessed and Modifiedtimes are
350 those of the time of the restore, not those of the originalfile.
351 The dates you will find in your restore log seem to be the original
353 - Volume "add"ed to Pool gets recycled in first use. VolBytes=0
354 - If a tape is recycled while it is mounted, Stanislav Tvrudy must do an
355 additional mount to deblock the job.
357 bscan: sql_update.c:65 UPDATE File SET MD5='Ij+5kwN6TFIxK+8l8+/I+A' WHERE FileId=0
358 bscan: bscan.c:1074 Could not add MD5/SHA1 to File record. ERR=sql_update.c:65 Update problem: affected_rows=0
359 - Do scheduling by UTC using gmtime_r() in run_conf, scheduler, and
360 ua_status.!!! Thanks to Alan Brown for this tip.
361 - Look at updating Volume Jobs so that Max Volume Jobs = 1 will work
362 correctly for multiple simultaneous jobs.
363 - Correct code so that FileSet MD5 is calculated for < and | filename
365 - Mark Volume in error on error from WEOF.
366 - Implement the Media record flag that indicates that the Volume does disk
368 - Implement VolAddr, which is used when Volume is addressed like a disk,
369 and form it from VolFile and VolBlock.
370 - Make multiple restore jobs for multiple media types specifying
371 the proper storage type.
372 - Implement MediaType keyword in bsr?
373 - Fix fast block rejection (stored/read_record.c:118). It passes a null
374 pointer (rec) to try_repositioning().
375 - Look at extracting Win data from BackupRead.
376 - Having dashes in filenames apparently creates problems for restore
377 by filename??? hard to believe.
378 - Implement RestoreJobRetention? Maybe better "JobRetention" in a Job,
379 which would take precidence over the Catalog "JobRetention".
380 - Implement Label Format in Add and Label console commands.
381 - Possibly up network buffers to 65K. Put on variable.
382 - Put email tape request delays on one or more variables. User wants
383 to cancel the job after a certain time interval. Maximum Mount Wait?
384 - Job, Client, Device, Pool, or Volume?
385 Is it possible to make this a directive which is *optional* in multiple
386 resources, like Level? If so, I think I'd make it an optional directive
387 in Job, Client, and Pool, with precedence such that Job overrides Client
388 which in turn overrides Pool.
389 - Print a message when a job starts if the conf file is not current.
391 - Spooling ideas taken from Volker Sauer's and other's emails:
392 > IMHO job spooling should be turned on
398 > where and 2) overrides 1) and 3) is independent.
400 Yes, this is the minimum that I think is necessary.
404 > It should be switched by job, because the job that backs up the machine
405 > with the bacula-sd on doesn't need spooling.
406 > It should be switched by schedule, because for full-backups I don't need
407 > spooling, so I can switch it off (because the network faster then the
410 True, with the exception that if you have enough disk spool space,
411 and you want to run concurrent jobs, spooling can eliminate the block
412 interleaving restore inefficiencies.
414 > And you should be able to turn it of by sd for sd-machines with low disk
415 > capacity or if you just don't need or want this feature.
418 > - definitly the possibility for multipe spool direcories
420 Having multiple directories is no problem -- having different maximum
421 sizes creates specification problems. At some point, I will probably
422 have a common SD pool of spool directories as well as a set of
423 private spool directories for each device. The first implementation
424 will be a set of private spool directories for each device since
425 managing a global pool with a bunch of threads writing into the same
426 directory is *much* more complicated and prone to error.
428 > - the ability to spool parts of a backup (not the whole client)
430 This may change in the future, but for the moment, it will spool
431 either to a job high water mark, or until the directory is full
432 (reaches max spool size or I/O error). It will then write to tape,
433 truncate the spool file, and begin spooling again.
435 > - spooling while writing to tape
437 Not within a job, but yes, if you run concurrent jobs -- each is a
438 different thread. Within a job could be a feature, but *much* later.
440 > - parallel spooling (like parallel jobs/ concurrent jobs) of clients
442 Yes, this is one of my main motivations for doing it (aside from
443 eliminating tape "shoe shine" during incremental backups.
445 > - flushing a backup that only went to disk (like amflush in amanda)
447 This will be a future feature, since spooling is different from backing
448 up to disk. The future feature will be "migration" which will move a job
449 from one backup Volume to another.
451 - New Storage specifications:
452 Passed to SD as a sort of BSR record called Storage Specification
456 MediaType -> Next MediaType
458 Device -> Next Device
459 Write Copy Resource that makes a copy of a resource.
461 Allow multiple Storage specifications
469 Allow Multiple Pool specifications (note, Pool currently
471 Allow Multiple MediaType specifications
472 Allow Multiple Device specifications
473 Perhaps keep this in a single SSR
474 Tie a Volume to a specific device by using a MediaType that
475 is contained in only one device.
476 In SD allow Device to have Multiple MediaTypes
479 - Look at www.nu2.nu/pebuilder as a helper for full windows
481 Ideas from Jerry Scharf:
482 First let's point out some big pluses that bacula has for this
484 more importantly it's active. Thank you so much for that
485 even more important, it's not flaky
486 it has an open access catalog, opening many possibilities
487 it's pushing toward heterogeneous systems capability
489 I don't remember an include file directive for config files
490 (not filesets, actual config directives)
491 can you check the configs without starting the daemon?
492 some warnings about possible common mistakes
494 doing the testing and blessing of concurrent backup writes
495 this is absolutely necessary in the enterprise
496 easy user recovery GUI with full access checking
497 Macintosh file client
498 macs are an interesting niche, but I fear a server is a rathole
499 working bare iron recovery for windows
500 much better handling on running config changes
501 thinking through the logic of what happens to jobs in progress
502 the option for inc/diff backups not reset on fileset revision
503 a) use both change and inode update time against base time
504 b) do the full catalog check (expensive but accurate)
505 sizing guide (how much system is needed to back up N systems/files)
506 consultants on using bacula in building a disaster recovery system
508 or how to get at fancy things that one could do with bacula
509 logwatch code for bacula logs (or similar)
510 linux distro inclusion of bacula (brings good and bad, but necessary)
511 win2k/XP server capability (icky but you asked)
512 support for Oracle database ??
514 - Look at adding SQL server and Exchange support for Windows.
515 - Restore: Enter Filename: 'C:/Documents and Settings/Comercial/My
516 Documents/MOP/formulário de registro BELAS ARTES.doc' causes Bacula to
518 - Each DVD-RAM disk would be a volume, just like each tape is
519 a volume. It's a 4.7GB media with random access, but there's nothing about
520 it that I can see that makes it so different than a tape from bacula's
521 perspective. Why couldn't I back up to a bare floppy as a volume (ignoring
523 - Make dev->file and dev->block_num signed integers so that -1 can
524 be an invalid value which happens with BSR.
525 - Create VolAddr for disk files in place of VolFile and VolBlock. This
526 is needed to properly specify ranges.
527 - Print bsmtp output to job report so that problems will be seen.
528 - Pass the number of files to be restored to the FD for reporting
529 - Add progress of files/bytes to SD and FD.
530 - Don't continue Restore if no files selected.
531 - Print warning message if FileId > 4 billion
532 - do a "messages" before the first prompt in Console
533 - Add a date and time stamp at the beginning of every line in the
534 Job report (Volker Sauer).
535 - Client does not show busy during Estimate command.
536 - Implement Console mtx commands.
537 - Add a default DB password to MySQL.
538 GRANT all privileges ON bacula.* TO bacula@localhost IDENTIFIED BY
541 - Implement a Mount Command and an Unmount Command where
542 the users could specify a system command to be performed
543 to do the mount, after which Bacula could attempt to
544 read the device. This is for Removeable media such as a CDROM.
545 - Most likely, this mount command would be invoked explicitly
546 by the user using the current Console "mount" and "unmount"
547 commands -- the Storage Daemon would do the right thing
548 depending on the exact nature of the device.
549 - As with tape drives, when Bacula wanted a new removable
550 disk mounted, it would unmount the old one, and send a message
551 to the user, who would then use "mount" as described above
552 once he had actually inserted the disk.
553 - Implement dump/print label to UA
554 - Implement disk spooling. Two parts: 1. Spool to disk then
555 immediately to tape to speed up tape operations. 2. Spool to
556 disk only when the tape is full, then when a tape is hung move
558 - Scratch Pool where the volumes can be re-assigned to any Pool.
559 - bextract is sending everything to the log file ****FIXME****
560 - Add Progress command that periodically reports the progress of
562 - Restrict characters permitted in a Resource name, and don't permit
564 - Allow multiple Storage specifications (or multiple names on
565 a single Storage specification) in the Job record. Thus a job
566 can be backed up to a number of storage devices.
567 - Implement some way for the File daemon to contact the Director
568 to start a job or pass its DHCP obtained IP number.
569 - Implement multiple Consoles.
570 - Implement a query tape prompt/replace feature for a console
572 Two jobs ready to go, first one blocked waiting for media
573 Cancel 2nd job ("waiting execution" one)
576 - Copy console @ code to gnome2-console
577 - Make AES the only encryption algorithm see
578 http://csrc.nist.gov/CryptoToolkit/aes/). It's
579 an officially adopted standard, has survived peer
580 review, and provides keys up to 256 bits.
581 - Think about how space could be freed up on a tape -- perhaps this
582 is a Merge or Compact feature that is needed.
583 - Modify FileSet, did not upgrade the current Increment job, but
584 waited for the next job to be upgraded.
585 - Take a careful look at SetACL http://setacl.sourceforge.net
586 - Implement a where command for the tree telling where a file
588 - Take a careful look at Level for the estimate command, maybe make
589 it a command line option.
590 - Add Volume name to "I cannot write on this volume because"
591 - Make tree walk routines like cd, ls, ... more user friendly
592 by handling spaces better.
593 - Write your PID file and chown root:wheel before drop.
594 - Make sure there is no symlink in a file before creating a
596 - Look at mktemp or mkstemp(3).
597 mktemp and mkstemp create files with predictable names too. That's
598 not the vulnerability. The vulnerability is in creating files without
599 using the O_EXCL flag, which means "only create this file if it doesn't
600 exist, including if the file is a dangling symlink."
602 It is *NOT* enough to do the equivalent of
604 if doesn't exist $filename
605 then create $filename
607 because between the test and the create another process could have
608 gotten the CPU and created the file. You must use atomic functions
609 (those that don't get interrupted by other processes) and O_EXCL is
610 the only way for this particular example.
611 - Automatically create pools, but instead of looking for what
612 in in Job records, walk through the pool resources.
613 - Check and double check tree code, why does it take so long?
614 - Add device name to "Current Volume not acceptable because ..."
615 - Make sure that Bacula rechecks the tape after the 20 min wait.
616 - Set IO_NOWAIT on Bacula TCP/IP packets.
617 - Try doing a raw partition backup and restore by mounting a
620 Yes, it would allow to highly automatic the request for new tapes. If a
621 tape is empty, bacula reads the barcodes (native or simulated), and if
622 an unused tape is found, it runs the label command with all the
623 necessary parameters.
625 By the way can bacula automatically "move" an empty/purged volume say
626 in the "short" pool to the "long" pool if this pool runs out of volume
628 - Eliminate orphaned jobs: dbcheck, normal pruning, delete job command.
629 Hm. Well, there are the remaining orphaned job records:
631 | 105 | Llioness Save | 0000-00-00 00:00:00 | B | D | 0 | 0 | f |
632 | 110 | Llioness Save | 0000-00-00 00:00:00 | B | I | 0 | 0 | f |
633 | 115 | Llioness Save | 2003-09-10 02:22:03 | B | I | 0 | 0 | A |
634 | 128 | Catalog Save | 2003-09-11 03:53:32 | B | I | 0 | 0 | C |
635 | 131 | Catalog Save | 0000-00-00 00:00:00 | B | I | 0 | 0 | f |
637 As you can see, three of the five are failures. I already deleted the
638 one restore and one other failure using the by-client option. Deciding
639 what is an orphaned job is a tricky problem though, I agree. All these
640 records have or had 0 files/ 0 bytes, except for the restore. With no
641 files, of course, I don't know of the job ever actually becomes
642 associated with a Volume.
644 (I'm not sure if this is documented anywhere -- what are the meanings of
645 all the possible JobStatus codes?)
647 Looking at my database, it appears to me as though all the "orphaned"
648 jobs fit into one of two categories:
650 1) The Job record has a StartTime but no EndTime, and the job is not
653 2) The Job record has an EndTime, indicating that it completed, but
654 it has no associated JobMedia record.
657 This does suggest an approach. If failed jobs (or jobs that, for some
658 other reason, write no files) are associated with a volume via a
659 JobMedia record, then they should be purged when the associated volume
660 is purged. I see two ways to handle jobs that are NOT associated with a
663 1) purge them automatically whenever any volume is manually purged;
665 2) add an option to the purge command to manually purge all jobs with
666 no associated volume.
668 I think Restore jobs also fall into category 2 above .... so one might
669 want to make that "The Job record has an EndTime,, but no associated
670 JobMedia record, and is not a Restore job."
671 - make "btape /tmp" work.
672 - Make sure a rescheduled job is properly reported by status.
673 - Walk through the Pool records rather than the Job records
674 in dird.c to create/update pools.
675 - What to do about "list files job=xxx".
676 - When job rescheduled, status gives is waiting for Client Rufus
677 to connect to Storage File. Dir needs to inform SD that job
679 - Make Dmsg look at global before calling subroutine.
680 - Enable trace output at runtime for Win32
681 - Available volumes for autochangers (see patrick@baanboard.com 3 Sep 03
682 and 4 Sep) scan slots.
683 - Get and test MySQL 4.0
684 - Do a complete audit of all pthreads_mutex, cond, ... to ensure that
685 any that are dynamically initialized are destroyed when no longer used.
686 - Look at how fuser works and /proc/PID/fd that is how Nic found the
687 file descriptor leak in Bacula.
688 - Implement WrapCounters in Counters.
689 - Turn on SIGHUP in dird.c and test.
690 - Use system dependent calls to get more precise info on tape errors.
691 - Add heartbeat from FD to SD if hb interval expires.
692 - Suppress read error on blank tape when doing a label.
693 - Can we dynamically change FileSets?
694 - If pool specified to label command and Label Format is specified,
695 automatically generate the Volume name.
696 - Why can't SQL do the filename sort for restore?
697 - Look at libkse (man kse) for FreeBSD threading.
698 - Look into Microsoft Volume Shadowcopy Service VSS for backing
699 up system state components (Active Directory, System Volume, ...)
700 - Add ExhautiveRestoreSearch
701 - Look at the possibility of loading only the necessary
702 data into the restore tree (i.e. do it one directory at a
703 time as the user walks through the tree).
704 - Possibly use the hash code if the user selects all for a restore command.
705 - Orphaned Dir buffer at parse_conf.c:373 => store_dir
706 - Fix "restore all" to bypass building the tree.
707 - Prohibit backing up archive device (findlib/find_one.c:128)
708 - Implement Release Device in the Job resource to unmount a drive.
709 - Implement Acquire Device in the Job resource to mount a drive,
710 be sure this works with admin jobs so that the user can get
711 prompted to insert the correct tape. Possibly some way to say to
712 run the job but don't save the files.
713 - Implement FileOptions (see end of this document)
714 - Make things like list where a file is saved case independent for
717 - Bacula needs to propagate SD errors.
718 > > cluster-dir: Start Backup JobId 252, Job=REUTERS.2003-08-11_15.04.12
719 > > prod4-sd: REUTERS.2003-08-11_15.04.12 Error: Write error on device
720 /dev/nst0. ERR=Input/output error.
721 > > prod4-sd: REUTERS.2003-08-11_15.04.12 Error: Re-read of last block failed.
722 Last block=5162 Current block=5164.
723 > > prod4-sd: End of medium on Volume "REU007" Bytes=16,303,521,933
725 - Use autochanger to handle multiple devices.
726 - Add SuSE install doc to list.
727 - Check and rechedk "Invalid block number"
728 - Make bextract release the drive properly between tapes
729 so that an autochanger can be made to work.
730 - User wants to NOT backup up certain big files (email files).
731 - Maybe remove multiple simultaneous devices code in SD.
732 - On Windows with very long path names, it may be impossible to create
733 a file (and thus restore it) because the total length is too long.
734 We must cd into the directory then create the file without the
736 - lstat() is not going to work on Win32 for testing date.
737 - Implement a Recycle command
738 - Add client name to cram-md5 challenge so Director can immediately
739 verify if it is the correct client.
740 - Add JobLevel in FD status (but make sure it is defined).
741 - Audit all UA commands to ensure that we always prompt where possible.
742 - Check Jmsg in bnet, may not work, must dup bsock.
743 - Suppress Job Name in Jmsg for console
744 - Create Pools that are referenced in a Run statement at startup if possible.
745 - Use runbeforejob to unload, then reload a volume previously used,
746 then the next job run gets an error reading the drive.
747 - Make bootstrap filename unique.
748 - Test a second language e.g. french.
749 - Start working on Base jobs.
750 - Make "make binary-release" work from any directory.
751 - Implement UnsavedFiles DB record.
752 - Implement argc/argv for daemon command line scanning using table driven
754 - Implement table driven single argc/argv scanner to pickup all arguments.
755 Much like xxx_conf.c scan table.
756 keyword, handler(store_routine), store_address, code, flags, default.
757 - From Phil Stracchino:
758 It would probably be a per-client option, and would be called
759 something like, say, "Automatically purge obsoleted jobs". What it
760 would do is, when you successfully complete a Differential backup of a
761 client, it would automatically purge all Incremental backups for that
762 client that are rendered redundant by that Differential. Likewise,
763 when a Full backup on a client completed, it would automatically purge
764 all Differential and Incremental jobs obsoleted by that Full backup.
765 This would let people minimize the number of tapes they're keeping on
766 hand without having to master the art of retention times.
767 - Implement a M_SECURITY message class.
768 - When doing a Backup send all attributes back to the Director, who
769 would then figure out what files have been deleted.
770 - Currently in mount.c:236 the SD simply creates a Volume. It should have
771 explicit permission to do so. It should also mark the tape in error
772 if there is an error.
773 - Make sure all restore counters are working correctly in the FD.
774 - SD Bytes Read is wrong.
775 - Look at ALL higher level routines that call block.c to be sure
776 they don't expect something in errmsg.
777 - Investigate doing RAW backup of Win32 partition.
778 - Add thread specific data to hold the jcr -- send error messages from
779 low level routines by accessing it and using Jmsg().
780 - Cancel waiting for Client connect in SD if FD goes away.
782 - Examine Bare Metal restore problem (a FD crash exists somewhere ...).
783 - Implement timeout in response() when it should come quickly.
784 - Implement console @echo command.
785 - Implement a Slot priority (loaded/not loaded).
786 - Implement "vacation" Incremental only saves.
787 - Implement single pane restore (much like the Gftp panes).
788 - Implement Automatic Mount even in operator wait.
789 - Implement create "FileSet"?
790 - Fix watchdog pthread crash on Win32 (this is pthread_kill() Cygwin bug)
791 - Implement "scratch pool" where tapes are defined and can be
792 taken by any pool that needs them.
793 - Implement restore "current system", but take all files without
794 doing selection tree -- so that jobs without File records can
796 - Add prefixlinks to where or not where absolute links to FD.
797 - Issue message to mount a new tape before the rewind.
798 - Simplified client job initiation for portables.
799 - If SD cannot open a drive, make it periodically retry.
800 - Add more of the config info to the tape label.
802 - If tape is marked read-only, then try opening it read-only rather than
803 failing, and remember that it cannot be written.
804 - Refine SD waiting output:
805 Device is being positioned
806 > Device is being positioned for append
807 > Device is being positioned to file x
809 - Figure out some way to estimate output size and to avoid splitting
810 a backup across two Volumes -- this could be useful for writing CDROMs
811 where you really prefer not to have it split -- not serious.
812 - Have SD compute MD5 or SHA1 and compare to what FD computes.
813 - Make VolumeToCatalog calculate an MD5 or SHA1 from the
814 actual data on the Volume and compare it.
815 - Implement Bacula plugins -- design API
816 - Make bcopy read through bad tape records.
817 - Program files (i.e. execute a program to read/write files).
818 Pass read date of last backup, size of file last time.
819 - Add Signature type to File DB record.
820 - CD into subdirectory when open()ing files for backup to
821 speed up things. Test with testfind().
822 - Priority job to go to top of list.
823 - Why are save/restore of device different sizes (sparse?) Yup! Fix it.
824 - Implement some way for the Console to dynamically create a job.
825 - Restore to a particular time -- e.g. before date, after date.
826 - Solaris -I on tar for include list
827 - Need a verbose mode in restore, perhaps to bsr.
828 - bscan without -v is too quiet -- perhaps show jobs.
829 - Add code to reject whole blocks if not wanted on restore.
830 - Check if we can increase Bacula FD priorty in Win2000
831 - Make sure the MaxVolFiles is fully implemented in SD
832 - Check if both CatalogFiles and UseCatalog are set to SD.
833 - Figure out how to do a bare metal Windows restore
834 - Possibly add email to Watchdog if drive is unmounted too
835 long and a job is waiting on the drive.
836 - Restore program that errs in SD due to no tape, reports
837 OK incorrectly in output.
838 - After unmount, if restore job started, ask to mount.
839 - Convert all %x substitution variables, which are hard to remember
840 and read to %(variable-name). Idea from TMDA.
841 - Remove NextId for SQLite. Optimize.
842 - Move all SQL statements into a single location.
843 - Add UA rc and history files.
844 - put termcap (used by console) in ./configure and
845 allow -with-termcap-dir.
846 - Fix Autoprune for Volumes to respect need for full save.
847 - Fix Win32 config file definition name on /install
848 - Compare tape to Client files (attributes, or attributes and data)
849 - Make all database Ids 64 bit.
850 - Write an applet for Linux.
851 - Allow console commands to detach or run in background.
852 - Fix status delay on storage daemon during rewind.
853 - Add SD message variables to control operator wait time
854 - Maximum Operator Wait
855 - Minimum Message Interval
856 - Maximum Message Interval
857 - Send Operator message when cannot read tape label.
858 - Verify level=Volume (scan only), level=Data (compare of data to file).
859 Verify level=Catalog, level=InitCatalog
861 - Add keyword search to show command in Console.
862 - Events : tape has more than xxx bytes.
863 - Complete code in Bacula Resources -- this will permit
864 reading a new config file at any time.
865 - Handle ctl-c in Console
866 - Implement script driven addition of File daemon to config files.
867 - Think about how to make Bacula work better with File (non-tape) archives.
868 - Write Unix emulator for Windows.
869 - Put memory utilization in Status output of each daemon
870 if full status requested or if some level of debug on.
871 - Make database type selectable by .conf files i.e. at runtime
872 - Set flag for uname -a. Add to Volume label.
873 - Implement throttled work queue.
874 - Restore files modified after date
875 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
876 - Implement Restore FileSet=
877 - Create a protocol.h and protocol.c where all protocol messages
879 - Remove duplicate fields from jcr (e.g. jcr.level and jcr.jr.Level, ...).
880 - Timout a job or terminate if link goes down, or reopen link and query.
881 - Concept of precious tapes (cannot be reused).
882 - Make bcopy copy with a single tape drive.
883 - Permit changing ownership during restore.
886 > My suggestion: Add a feature on the systray menu-icon menu to request
887 > an immediate backup now. This would be useful for laptop users who may
888 > not be on the network when the regular scheduled backup is run.
890 > My wife's suggestion: Add a setting to the win32 client to allow it to
891 > shut down the machine after backup is complete (after, of course,
892 > displaying a "System will shut down in one minute, click here to cancel"
893 > warning dialog). This would be useful for sites that want user
894 > woorkstations to be shut down overnight to save power.
897 - Autolabel should be specified by DIR instead of SD.
900 - AutoScan (check checksum of tape)
901 - Format command = "format /dev/nst0"
905 - Seek resolution (usually corresponds to buffer size)
906 - EODErrorCode=ENOSPC or code
907 - Partial Read error code
908 - Partial write error code
909 - Nonformatted read error
910 - Nonformatted write error
911 - WriteProtected error
915 - IgnoreCloseErrors=yes
925 - FD sends unsaved file list to Director at end of job (see
927 - File daemon should build list of files skipped, and then
928 at end of save retry and report any errors.
929 - Write a Storage daemon that uses pipes and
930 standard Unix programs to write to the tape.
932 - Need something that monitors the JCR queue and
933 times out jobs by asking the deamons where they are.
934 - Enhance Jmsg code to permit buffering and saving to disk.
935 - device driver = "xxxx" for drives.
937 - Ensure that /dev/null works
938 - Need report class for messages. Perhaps
939 report resource where report=group of messages
940 - enhance scan_attrib and rename scan_jobtype, and
941 fill in code for "since" option
942 - Director needs a time after which the report status is sent
943 anyway -- or better yet, a retry time for the job.
944 - Don't reschedule a job if previous incarnation is still running.
945 - Some way to automatically backup everything is needed????
946 - Need a structure for pending actions:
948 - termination status (part of buffered msgs?)
950 Read, Write, Clean, Delete
951 - Login to Bacula; Bacula users with different permissions:
952 owner, group, user, quotas
953 - Store info on each file system type (probably in the job header on tape.
954 This could be the output of df; or perhaps some sort of /etc/mtab record.
956 ========= ideas ===============
957 From: "Jerry K. Schieffer" <jerry@skylinetechnology.com>
958 To: <kern@sibbald.com>
959 Subject: RE: [Bacula-users] future large programming jobs
960 Date: Thu, 26 Feb 2004 11:34:54 -0600
962 I noticed the subject thread and thought I would offer the following
963 merely as sources of ideas, i.e. something to think about, not even as
964 strong as a request. In my former life (before retiring) I often
965 dealt with backups and storage management issues/products as a
966 developer and as a consultant. I am currently migrating my personal
967 network from amanda to bacula specifically because of the ability to
968 cross media boundaries during storing backups.
969 Are you familiar with the commercial product called ADSM (I think IBM
970 now sells it under the Tivoli label)? It has a couple of interesting
971 ideas that may apply to the following topics.
973 1. Migration: Consider that when you need to restore a system, there
974 may be pressure to hurry. If all the information for a single client
975 can eventually end up on the same media (and in chronological order),
976 the restore is facillitated by not having to search past information
977 from other clients. ADSM has the concept of "client affinity" that
978 may be associated with it's storage pools. It seems to me that this
979 concept (as an optional feature) might fit in your architecture for
982 ADSM also has the concept of defining one or more storage pools as
983 "copy pools" (almost mirrors, but only in the sense of contents).
984 These pools provide the ability to have duplicte data stored both
985 onsite and offsite. The copy process can be scheduled to be handled
986 by their storage manager during periods when there is no backup
987 activity. Again, the migration process might be a place to consider
988 implementing something like this.
991 > It strikes me that it would be very nice to be able to do things
993 > have the Job(s) backing up the machines run, and once they have all
994 > completed, start a migration job to copy the data from disks Volumes
996 > a tape library and then to offsite storage. Maybe this can already
998 > done with some careful scheduling and Job prioritzation; the events
999 > mechanism described below would probably make it very easy.
1001 This is the goal. In the first step (before events), you simply
1003 the Migration to tape later.
1005 2. Base jobs: In ADSM, each copy of each stored file is tracked in
1006 the database. Once a file (unique by path and metadata such as dates,
1007 size, ownership, etc.) is in a copy pool, no more copies are made. In
1008 other words, when you start ADSM, it begins like your concept of a
1009 base job. After that it is in the "incremental" mode. You can
1010 configure the number of "generations" of files to be retained, plus a
1011 retention date after which even old generations are purged. The
1012 database tracks the contents of media and projects the percentage of
1013 each volume that is valid. When the valid content of a volume drops
1014 below a configured percentage, the valid data are migrated to another
1015 volume and the old volume is marked as empty. Note, this requires
1016 ADSM to have an idea of the contents of a client, i.e. marking the
1017 database when an existing file was deleted, but this would solve your
1018 issue of restoring a client without restoring deleted files.
1020 This is pretty far from what bacula now does, but if you are going to
1021 rip things up for Base jobs,.....
1022 Also, the benefits of this are huge for very large shops, especially
1023 with media robots, but are a pain for shops with manual media
1027 > Base jobs sound pretty useful, but I'm not dying for them.
1029 Nobody is dying for them, but when you see what it does, you will die
1032 3. Restoring deleted files: Since I think my comments in (2) above
1033 have low probability of implementation, I'll also suggest that you
1034 could approach the issue of deleted files by a mechanism of having the
1035 fd report to the dir, a list of all files on the client for every
1036 backup job. The dir could note in the database entry for each file
1037 the date that the file was seen. Then if a restore as of date X takes
1038 place, only files that exist from before X until after X would be
1039 restored. Probably the major cost here is the extra date container in
1040 each row of the files table.
1042 Thanks for "listening". I hope some of this helps. If you want to
1043 contact me, please send me an email - I read some but not all of the
1044 mailing list traffic and might miss a reply there.
1046 Please accept my compliments for bacula. It is doing a great job for
1047 me!! I sympathize with you in the need to wrestle with excelence in
1048 execution vs. excelence in feature inclusion.
1053 ==============================
1056 - Design at hierarchial storage for Bacula. Migration and Clone.
1057 - Implement FSM (File System Modules).
1058 - Audit M_ error codes to ensure they are correct and consistent.
1059 - Add variable break characters to lex analyzer.
1060 Either a bit mask or a string of chars so that
1061 the caller can change the break characters.
1062 - Make a single T_BREAK to replace T_COMMA, etc.
1063 - Ensure that File daemon and Storage daemon can
1064 continue a save if the Director goes down (this
1065 is NOT currently the case). Must detect socket error,
1066 buffer messages for later.
1067 - Enhance time/duration input to allow multiple qualifiers e.g. 3d2h
1068 - Add ability to backup to two Storage devices (two SD sessions) at
1069 the same time -- e.g. onsite, offsite.
1070 - Add the ability to consolidate old backup sets (basically do a restore
1071 to tape and appropriately update the catalog). Compress Volume sets.
1072 Might need to spool via file is only one drive is available.
1073 - Compress or consolidate Volumes of old possibly deleted files. Perhaps
1074 someway to do so with every volume that has less than x% valid
1078 Migration: Move a backup from one Volume to another
1079 Clone: Copy a backup -- two Volumes
1081 Bacula Migration is based on Jobs (apparently Networker is file by file).
1083 Migration triggered by:
1087 Highwater mark (keep total size)
1092 ======================================================
1094 It is somewhat like a Full save becomes an incremental since
1095 the Base job (or jobs) plus other non-base files.
1097 - A Base backup is same as Full backup, just different type.
1098 - New BaseFiles table that contains:
1100 BaseJobId - Base JobId referenced for this FileId (needed ???)
1101 JobId - JobId currently running
1102 FileId - File not backed up, exists in Base Job
1103 FileIndex - FileIndex from Base Job.
1104 i.e. for each base file that exists but is not saved because
1105 it has not changed, the File daemon sends the JobId, BaseId,
1106 FileId, FileIndex back to the Director who creates the DB entry.
1107 - To initiate a Base save, the Director sends the FD
1108 the FileId, and full filename for each file in the Base.
1109 - When the FD finds a Base file, he requests the Director to
1110 send him the full File entry (stat packet plus MD5), or
1111 conversely, the FD sends it to the Director and the Director
1112 says yes or no. This can be quite rapid if the FileId is kept
1113 by the FD for each Base Filename.
1114 - It is probably better to have the comparison done by the FD
1115 despite the fact that the File entry must be sent across the
1117 - An alternative would be to send the FD the whole File entry
1118 from the start. The disadvantage is that it requires a lot of
1119 space. The advantage is that it requires less communications
1121 - The Job record must be updated to indicate that one or more
1123 - At end of Job, FD returns:
1124 1. Count of base files/bytes not written to tape (i.e. matches)
1125 2. Count of base file that were saved i.e. had changed.
1126 - No tape record would be written for a Base file that matches, in the
1127 same way that no tape record is written for Incremental jobs where
1128 the file is not saved because it is unchanged.
1129 - On a restore, all the Base file records must explicitly be
1130 found from the BaseFile tape. I.e. for each Full save that is marked
1131 to have one or more Base Jobs, search the BaseFile for all occurrences
1133 - An optimization might be to make the BaseFile have:
1139 This would avoid the need to explicitly fetch each File record for
1140 the Base job. The Base Job record will be fetched to get the
1141 VolSessionId and VolSessionTime.
1142 =========================================================
1145 ==========================================================
1147 For each Incremental job that is run, there may be files that
1148 were found but not saved because they were locked (this applies
1149 only to Windows). Such a system could send back to the Director
1150 a list of Unsaved files.
1152 - New UnSavedFiles table that contains:
1156 - Then in the next Incremental job, the list of Unsaved Files will be
1157 feed to the FD, who will ensure that they are explicitly chosen even
1158 if standard date/time check would not have selected them.
1159 =============================================================
1162 Done: (see kernsdone for more)
1165 - John's Full save failed with 1.32c FD and 1.31 Dir no FD status,
1166 and no error message.
1167 - Add fd and st as Console keywords.
1168 - Recycling volume with a Slot requires an operator intervention:
1169 rufus-dir: Start Backup JobId 18, Job=kernsave.2003-11-01_21.23.52
1170 rufus-dir: Pruned 1 Job on Volume Vol01 from catalog.
1171 rufus-dir: There are no Jobs associated with Volume Vol01. Marking it purged.
1172 rufus-dir: Recycled volume "Vol01"
1173 rufus-sd: Please mount Volume "Vol01" on Storage Device "DDS-4" for Job kernsave.2003-11-01_21.23.52
1174 Use "mount" command to release Job.
1175 - Implement Dan's bacula script (email of 26 Oct).
1176 - Add JobName= to VerifyToCatalog so that all verifies can be done at the end.
1177 - Edit the Client/Storage name into authentication failure messages.
1178 - Fix packet too big problem. This is most likely a Windows TCP stack
1180 - Implement ClientRunBeforeJob and ClientRunAfterJob.
1181 - Implement forward spacing block/file: position_device(bsr) --
1182 just before read_block_from_device();
1185 Multiple drive autochanger data: see Alan Brown
1186 mtx -f xxx unloadStorage Element 1 is Already Full(drive 0 was empty)
1187 Unloading Data Transfer Element into Storage Element 1...source Element
1188 Address 480 is Empty
1190 (drive 0 was empty and so was slot 1)
1191 > mtx -f xxx load 15 0
1192 no response, just returns to the command prompt when complete.
1193 > mtx -f xxx status Storage Changer /dev/changer:2 Drives, 60 Slots ( 2 Import/Export )
1194 Data Transfer Element 0:Full (Storage Element 15 Loaded):VolumeTag = HX001
1195 Data Transfer Element 1:Empty
1196 Storage Element 1:Empty
1197 Storage Element 2:Full :VolumeTag=HX002
1198 Storage Element 3:Full :VolumeTag=HX003
1199 Storage Element 4:Full :VolumeTag=HX004
1200 Storage Element 5:Full :VolumeTag=HX005
1201 Storage Element 6:Full :VolumeTag=HX006
1202 Storage Element 7:Full :VolumeTag=HX007
1203 Storage Element 8:Full :VolumeTag=HX008
1204 Storage Element 9:Full :VolumeTag=HX009
1205 Storage Element 10:Full :VolumeTag=HX010
1206 Storage Element 11:Empty
1207 Storage Element 12:Empty
1208 Storage Element 13:Empty
1209 Storage Element 14:Empty
1210 Storage Element 15:Empty
1211 Storage Element 16:Empty....
1212 Storage Element 28:Empty
1213 Storage Element 29:Full :VolumeTag=CLNU01L1
1214 Storage Element 30:Empty....
1215 Storage Element 57:Empty
1216 Storage Element 58:Full :VolumeTag=NEX261L2
1217 Storage Element 59 IMPORT/EXPORT:Empty
1218 Storage Element 60 IMPORT/EXPORT:Empty
1220 Unloading Data Transfer Element into Storage Element 15...done
1222 (just to verify it remembers where it came from, however it can be
1223 overrriden with mtx unload {slotnumber} to go to any storage slot.)
1225 There needs to be a table of drive # to devices somewhere - If there are
1226 multiple changers or drives there may not be a 1:1 correspondance between
1227 changer drive number and system device name - and depending on the way the
1228 drives are hooked up to scsi busses, they may not be linearly numbered
1229 from an offset point either.something like
1231 Autochanger drives = 2
1232 Autochanger drive 0 = /dev/nst1
1233 Autochanger drive 1 = /dev/nst2
1234 IMHO, it would be _safest_ to use explicit mtx unload commands at all
1235 times, not just for multidrive changers. For a 1 drive changer, that's
1241 MTX's manpage (1.2.15):
1242 unload [<slotnum>] [ <drivenum> ]
1243 Unloads media from drive <drivenum> into slot
1244 <slotnum>. If <drivenum> is omitted, defaults to
1245 drive 0 (as do all commands). If <slotnum> is
1246 omitted, defaults to the slot that the drive was
1247 loaded from. Note that there's currently no way
1248 to say 'unload drive 1's media to the slot it
1249 came from', other than to explicitly use that
1250 slot number as the destination.AB
1256 undef# camcontrol devlist
1257 <WANGTEK 51000 SCSI M74H 12B3> at scbus0 target 2 lun 0 (pass0,sa0)
1258 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 0 (pass1,sa1)
1259 <ARCHIVE 4586XX 28887-XXX 4BGD> at scbus0 target 4 lun 1 (pass2)
1261 tapeinfo -f /dev/sg0 with a bad tape in drive 1:
1262 [kern@rufus mtx-1.2.17kes]$ ./tapeinfo -f /dev/sg0
1263 Product Type: Tape Drive
1265 Product ID: 'C5713A '
1267 Attached Changer: No
1268 TapeAlert[3]: Hard Error: Uncorrectable read/write error.
1269 TapeAlert[20]: Clean Now: The tape drive neads cleaning NOW.
1276 Medium Type: Not Loaded
1279 DataCompEnabled: yes
1280 DataCompCapable: yes
1281 DataDeCompEnabled: yes
1288 Handling removable disks
1290 From: Karl Cunningham <karlc@keckec.com>
1292 My backups are only to hard disk these days, in removable bays. This is my
1293 idea of how a backup to hard disk would work more smoothly. Some of these
1294 things Bacula does already, but I mention them for completeness. If others
1295 have better ways to do this, I'd like to hear about it.
1297 1. Accommodate several disks, rotated similar to how tapes are. Identified
1298 by partition volume ID or perhaps by the name of a subdirectory.
1299 2. Abort & notify the admin if the wrong disk is in the bay.
1300 3. Write backups to different subdirectories for each machine to be backed
1302 4. Volumes (files) get created as needed in the proper subdirectory, one
1304 5. When a disk is recycled, remove or zero all old backup files. This is
1305 important as the disk being recycled may be close to full. This may be
1306 better done manually since the backup files for many machines may be
1307 scattered in many subdirectories.
1312 - To pass Include 1 or two letter commands
1313 I Name Include name - first record
1314 B Name Base name - repeat
1317 E zzz Exclude expression (wild card)
1319 D "reader" Reader program
1320 T "writer" Writer program
1321 O Options In current commpressed format
1322 (compression, signature, onefs, recurse, sparse,
1323 replace, verify options, ...)
1325 B BaseName Start second option set any letter
1328 F Number Number of filenames to follow
1333 F Number Number of filenames to follow
1336 - Add regression of btape "fill"
1338 - Implement "label slots=1,2... pool=yyy barcodes"
1339 - Mark a non-removable disk Volume in error rather than asking sys op to
1341 - Implement Fixed storage LabelFormat test.
1342 - Add reporting in attr despooling.
1343 - Figure out how to use ssh or stunnel to protect Bacula communications.
1344 Add Dan's work to manual See ssl.txt
1347 - Implement fast tree insert (doubly linked list)
1348 - Disallow using Internal database
1349 - Feedback while the restore tree is being built.
1350 - Add detection of Whitebox release in ./configure
1351 - Add InChanger to "list media"
1352 - Implement second cut of SIGHUP in Dir
1353 (Implement resources on a single pointer)
1354 - Add regular expressions to FileSets (needs testing).
1355 - Test new despooling error recovery code when partition fills.
1356 - Sort Scheduled jobs status listing by start time.
1357 - Add priority to Scheduled jobs status listing.
1358 - Look at Win32 inc problem when new directory added files not saved???
1359 - Require resource names to be unique.
1360 - Make sure that all errors from libacl are printed.
1361 - Make FD run on Win95 if possible (not really possible).
1362 - Fix GROUP BYs in all SQL (done by Dan -- thanks)
1363 - Make entering multiple Storage names in Dir illegal.
1364 - Make btape release the drive during the "test" append.
1366 You have selected the following JobIds: 739,856,797
1367 Building directory tree for JobId 739 ...
1368 Building directory tree for JobId 856 ...
1369 Building directory tree for JobId 797 ...3
1370 Jobs inserted into the tree.
1371 - Fix errno handling in win32 compat routines.
1372 - Add better error codes to run_program (10000+)
1373 - Revisit and revise Disaster Recovery (fix SCSI and RAID
1375 - When passwords do not match, print message that points the
1377 - Do tape alerts -- see tapealert.txt
1378 - Find memory leaks in address_conf.c
1379 - Network order of port and ip addresses.
1380 - SetIP in dird/fd_cmds.c is broken.
1381 - don't use inet_addr() it is deprecated.
1382 - Fix code in filed.c for init_bsock();
1383 - Fix code in console.c for init_bsock();
1384 - Solve the termcap.h problem on Solaris configure.
1385 - Make Verify jobs require exclusive use of Volume as Restore
1387 - Fix restore ++++ that get intermingled with "Building directory tree"
1388 - Fix doc in dirdconf.html for @, where the file must
1389 have not a list of files, but
1392 - Test Win32 errno handling.
1393 - port 1.34.6-block.patch to 1.35
1394 - Look at Chris' patch for bscan bug.
1395 - Fix error handling in spooling both data and attribute.
1396 - Implement Ignore FileSet Change.
1397 - Doc new duration time input editing.
1398 - Bacula rescue CDROM implement isolinux
1399 - Make sure Qmsgs are dequeued by FD and SD.
1400 - Check if ACLs allocated at dird_conf.c:1214 are being properly
1402 - Test/doc Tape Alerts
1403 - Doc dbcheck eliminate orphaned clients.
1404 - Doc Phil's new delete job jobid scanning code.
1405 - Document that console commands can be abbreviated.
1406 - Doc update AllFromVol
1407 - Doc -p option in stored