4 Documentation to do: (a little bit at a time)
5 - Document running a test version.
6 - Document query file format.
7 - Document static linking
8 - Document how to automatically backup all local partitions
9 - Document problems with Verify and pruning.
10 - Document how to use multiple databases.
11 - Document Maximum File Size
12 - Add a section to the doc on Manual cycling of Volumes.
15 - Document default file backup
17 Testing to do: (painful)
18 - that ALL console command line options work and are always implemented
19 - blocksize recognition code.
20 - multiple simultaneous Volumes
23 - Fix Bare Metal restore problem.
24 - Fix "access not allowed" for backup of files on WinXP.
26 - Figure out how to use ssh or stunnel to protect Bacula communications.
29 - Implement a M_SECURITY message class.
30 - Allow multiple Storage specifications (or multiple names on
31 a single Storage specification) in the Job record. Thus a job
32 can be backed up to a number of storage devices.
33 - Implement dump label to UA
34 - Add prefixlinks to where or not where absolute links to FD.
35 - Look at Python for a Bacula scripting language -- www.python.org
36 - Issue message to mount a new tape before the rewind.
37 - Simplified client job initiation for portables.
38 - If SD cannot open a drive, make it periodically retry.
39 - Implement LabelTemplate (at least first cut).
40 - Add more of the config info to the tape label.
41 - Implement a Mount Command and an Unmount Command where
42 the users could specify a system command to be performed
43 to do the mount, after which Bacula could attempt to
44 read the device. This is for Removeable media such as a CDROM.
45 - Most likely, this mount command would be invoked explicitly
46 by the user using the current Console "mount" and "unmount"
47 commands -- the Storage Daemon would do the right thing
48 depending on the exact nature of the device.
49 - As with tape drives, when Bacula wanted a new removable
50 disk mounted, it would unmount the old one, and send a message
51 to the user, who would then use "mount" as described above
52 once he had actually inserted the disk.
54 - Make some way so that if a machine is skipped because it is not up
55 that Bacula will continue retrying for a specified period of time --
57 - If tape is marked read-only, then try opening it read-only rather than
58 failing, and remember that it cannot be written.
59 - Refine SD waiting output:
60 Device is being positioned
61 > Device is being positioned for append
62 > Device is being positioned to file x
64 - Figure out some way to estimate output size and to avoid splitting
65 a backup across two Volumes -- this could be useful for writing CDROMs
66 where you really prefer not to have it split -- not serious.
67 - Add RunBeforeJob and RunAfterJob to the Client program.
68 - Have SD compute MD5 or SHA1 and compare to what FD computes.
69 - Make VolumeToCatalog calculate an MD5 or SHA1 from the
70 actual data on the Volume and compare it.
71 - Implement FileOptions (see end of this document)
72 - Implement Bacula plugins -- design API
73 - Make bcopy read through bad tape records.
74 - Fix read_record to handle multiple sessions.
75 - Program files (i.e. execute a program to read/write files).
76 Pass read date of last backup, size of file last time.
77 - Add Signature type to File DB record.
78 - Make Restore report an error if FD or SD term codes are not OK.
79 - CD into subdirectory when open()ing files for backup to
80 speed up things. Test with testfind().
81 - Priority job to go to top of list.
82 - Find out why Full saves run slower and slower (hashing?)
83 - Why are save/restore of device different sizes (sparse?) Yup! Fix it.
84 - Implement some way for the Console to dynamically create a job.
85 - Restore to a particular time -- e.g. before date, after date.
86 - Implement disk spooling
87 - Solaris -I on tar for include list
88 - Prohibit backing up archive device (findlib/find_one.c:128)
89 - Need a verbose mode in restore, perhaps to bsr.
90 - bscan without -v is too quiet -- perhaps show jobs.
91 - Add code to reject whole blocks if not wanted on restore.
92 - Start working on Base jobs.
93 - Check if we can increase Bacula FD priorty in Win2000
94 - Make sure the MaxVolFiles is fully implemented in SD
95 - Check if both CatalogFiles and UseCatalog are set to SD.
96 - Need return status on read_cb() from read_records(). Need multiple
97 records -- one per Job, maybe a JCR or some other structure with
99 - Figure out how to do a bare metal Windows restore
100 - Put system type returned by FD into catalog.
101 - Possibly add email to Watchdog if drive is unmounted too
102 long and a job is waiting on the drive.
103 - Use read_record.c in SD code.
104 - Why don't we get an error message from Win32 FD when bootstrap
105 file cannot be created for restore command?
106 - When Marking a file in Restore that is a hard link, also
107 mark the link so that the data will be reloaded.
108 - Restore program that errors in SD due to no tape reports
109 OK incorrectly in output.
110 - After unmount, if restore job started, ask to mount.
111 - Convert all %x substitution variables, which are hard to remember
112 and read to %(variable-name). Idea from TMDA.
113 - Add JobLevel in FD status (but make sure it is defined).
114 - Make Pool resource handle Counter resources.
115 - Remove NextId for SQLite. Optimize.
116 - Move all SQL statements into a single location.
117 - Add UA rc and history files.
118 - put termcap (used by console) in ./configure and
119 allow -with-termcap-dir.
120 - Enhance time and size scanning routines.
121 - Fix Autoprune for Volumes to respect need for full save.
122 - Fix Win32 config file definition name on /install
123 - No READLINE_SRC if found in alternate directory.
124 - Test a second language e.g. french.
125 - Compare tape to Client files (attributes, or attributes and data)
126 - Make all database Ids 64 bit.
127 - Write an applet for Linux.
128 - Add estimate to Console commands
129 - Find solution to blank filename (i.e. path only) problem.
130 - Implement new daemon communications protocol.
131 - Remove PoolId from Job table, it exists in Media.
132 - Allow console commands to detach or run in background.
133 - Fix status delay on storage daemon during rewind.
134 - Add SD message variables to control operator wait time
135 - Maximum Operator Wait
136 - Minimum Message Interval
137 - Maximum Message Interval
138 - Send Operator message when cannot read tape label.
139 - Verify level=Volume (scan only), level=Data (compare of data to file).
140 Verify level=Catalog, level=InitCatalog
142 - Add keyword search to show command in Console.
143 - Fix Win2000 error with no messages during startup.
144 - Events : tape has more than xxx bytes.
145 - Restrict characters permitted in a Resource name.
146 - Complete code in Bacula Resources -- this will permit
147 reading a new config file at any time.
148 - Handle ctl-c in Console
149 - Implement script driven addition of File daemon to config files.
150 - Think about how to make Bacula work better with File (non-tape) archives.
151 - Write Unix emulator for Windows.
152 - Implement new serialize subroutines
153 send(socket, "string", &Vol, "uint32", &i, NULL)
154 - Audit all UA commands to ensure that we always prompt where possible.
155 - If ./btape is called without /dev, assume argument is a Storage resource name.
156 - Put memory utilization in Status output of each daemon
157 if full status requested or if some level of debug on.
158 - Make database type selectable by .conf files i.e. at runtime
159 - Set flag for uname -a. Add to Volume label.
160 - Implement throttled work queue.
161 - Check for EOT at ENOSPC or EIO or ENXIO (unix Pc)
162 - Restore files modified after date
163 - Restore file modified before date
164 - Emergency restore info:
166 - Backup working directory
168 - Restore -- do nothing but show what would happen
169 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
170 - Implement Restore FileSet=
171 - Create a protocol.h and protocol.c where all protocol messages
173 - Remove duplicate fields from jcr (e.g. jcr.level and jcr.jr.Level, ...).
174 - Timout a job or terminate if link goes down, or reopen link and query.
175 - Find general solution for sscanf size problems (as well
176 as sprintf. Do at run time?
177 - Concept of precious tapes (cannot be reused).
178 - Make bcopy copy with a single tape drive.
179 - Permit changing ownership during restore.
181 - Autolabel should be specified by DIR instead of SD.
182 - Find out how to get the system tape block limits, e.g.:
183 Apr 22 21:22:10 polymatou kernel: st1: Block limits 1 - 245760 bytes.
184 Apr 22 21:22:10 polymatou kernel: st0: Block limits 2 - 16777214 bytes.
187 - AutoScan (check checksum of tape)
188 - Format command = "format /dev/nst0"
192 - Seek resolution (usually corresponds to buffer size)
193 - EODErrorCode=ENOSPC or code
194 - Partial Read error code
195 - Partial write error code
196 - Nonformatted read error
197 - Nonformatted write error
198 - WriteProtected error
202 - IgnoreCloseErrors=yes
212 - FD sends unsaved file list to Director at end of job (see
214 - File daemon should build list of files skipped, and then
215 at end of save retry and report any errors.
216 - Write a Storage daemon that uses pipes and
217 standard Unix programs to write to the tape.
219 - Need something that monitors the JCR queue and
220 times out jobs by asking the deamons where they are.
221 - Enhance Jmsg code to permit buffering and saving to disk.
222 - device driver = "xxxx" for drives.
223 - restart: paranoid: read label fsf to
224 eom read append block, and go
225 super-paranoid: read label, read all files
226 in between, read append block, and go
227 verify: backspace, read append block, and go
228 permissive: same as above but frees drive
229 if tape is not valid.
231 - Ensure that /dev/null works
232 - Need report class for messages. Perhaps
233 report resource where report=group of messages
234 - enhance scan_attrib and rename scan_jobtype, and
235 fill in code for "since" option
236 - Director needs a time after which the report status is sent
237 anyway -- or better yet, a retry time for the job.
238 Don't reschedule a job if previous incarnation is still running.
239 - Some way to automatically backup everything is needed????
240 - Need a structure for pending actions:
242 - termination status (part of buffered msgs?)
244 Read, Write, Clean, Delete
245 - Login to Bacula; Bacula users with different permissions:
246 owner, group, user, quotas
247 - Store info on each file system type (probably in the job header on tape.
248 This could be the output of df; or perhaps some sort of /etc/mtab record.
251 - Design at hierarchial storage for Bacula. Migration and Clone.
252 - Implement FSM (File System Modules).
253 - Identify unchanged or "system" files and save them to a
254 special tape thus removing them from the standard
255 backup FileSet -- BASE backup.
256 - Heartbeat between daemons.
257 - Audit M_ error codes to ensure they are correct and consistent.
258 - Add variable break characters to lex analyzer.
259 Either a bit mask or a string of chars so that
260 the caller can change the break characters.
261 - Make a single T_BREAK to replace T_COMMA, etc.
262 - Ensure that File daemon and Storage daemon can
263 continue a save if the Director goes down (this
264 is NOT currently the case). Must detect socket error,
265 buffer messages for later.
266 - Enhance time/duration input to allow multiple qualifiers e.g. 3d2h
267 - Add ability to backup to two Storage devices (two SD sessions) at
268 the same time -- e.g. onsite, offsite.
269 - Add the ability to consolidate old backup sets (basically do a restore
270 to tape and appropriately update the catalog). Compress Volume sets.
271 Might need to spool via file is only one drive is available.
272 - Compress or consolidate Volumes of old possibly deleted files. Perhaps
273 someway to do so with every volume that has less than x% valid
278 Bacula Projects Roadmap
280 last update 5 January 2003
282 Item 1: Multiple simultaneous Jobs. (done)
283 Done -- Restore part needs better implementation to work correctly
285 What: Permit multiple simultaneous jobs in Bacula.
287 Why: An enterprise level solution needs to go fast without the
288 need for the system administrator to carefully tweak
289 timing. Based on the benchmarks, during a full
290 backup, NetWorker typically hit 10 times the bandwidth to
291 the tape compared to Bacula--largely. This is probably due to
292 running parallel jobs and multi-threaded filling of buffers
293 and writing them to tape. This should also make things work
294 better when you have a mix of fast and slow machines backing
297 Notes: Bacula was designed to run multiple simultaneous jobs. Thus
298 implementing this is a matter of some small cleanups and
302 Item 2: Make the Storage daemon use intermediate file storage to buffer data.
303 Deferred -- not necessary yet.
305 What: If data is coming into the SD too fast, buffer it to
306 disk if the user has configured this option.
308 Why: This would be nice, especially if it more or less falls out
309 when implementing (1) above. If not, it probably should not
310 be given a high priority because fundamentally the backup time
311 is limited by the tape bandwidth. Even though you may finish a
312 client job quicker by spilling to disk, you still have to
313 eventually get it onto tape. If intermediate disk buffering
314 allows us to improve write bandwidth to tape, it may make
317 Notes: Whether or not this is implemented will depend upon performance
318 testing after item 1 is implemented.
321 Item 3: Write the bscan program -- also write a bcopy program.
324 What: Write a program that reads a Bacula tape and puts all the
325 appropriate data into the catalog. This allows recovery
326 from a tape that is no longer in the database, or it allows
327 re-creation of a database if lost.
329 Why: This is a fundamental robustness and disaster recovery tool
330 which will increase the comfort level of a sysadmin
331 considering adopting Bacula.
333 Notes: A skeleton of this program already exists, but much work
334 needs to be done. Implementing this will also make apparent
335 any deficiencies in the current Bacula tape format.
338 Item 4: Implement Base jobs.
340 What: A base job is sort of like a Full save except that you
341 will want the FileSet to contain only files that are unlikely
342 to change in the future (i.e. a snapshot of most of your
343 system after installing it). After the base job has been run,
344 when you are doing a Full save, you can specify to exclude
345 all files saved by the base job that have not been modified.
347 Why: This is something none of the competition does, as far as we know
348 (except BackupPC, which is a Perl program that saves to disk
349 only). It is big win for the user, it makes Bacula stand out
350 as offering a unique optimization that immediately saves time
353 Notes: Big savings in tape usage. Will require more resources because
354 the e. DIR must send FD a list of files/attribs, and the FD must
355 search the list and compare it for each file to be saved.
358 Item 5: Implement Label templates
360 What: This is a mechanism whereby Bacula can automatically create
361 a tape label for new tapes according to a detailed specification
362 provided by the user.
364 Why: It is a major convenience item for folks who use automated label
367 Notes: Bacula already has a working form of automatic tape label
368 creation, but it is very crude. The design for the complete
369 tape labeling project is already documented in the manual.
372 Item 6: Write a regression script.
375 What: This is an automatic script that runs and tests as many features
376 of Bacula as possible. The output is compared to previous
377 versions of Bacula and any differences are reported.
379 Why: This is an enormous help in preventing introduction of new
380 errors in parts of the program that already work correctly.
382 Notes: This probably should be ranked higher, it's something the typical
383 user doesn't see. Depending on how it's implemented, it may
384 make sense to defer it until the archival tape format and
385 user interface mature.
388 Item 7: GUI for interactive restore
389 Item 8: GUI for interactive backup
391 What: The current interactive restore is implemented with a tty
392 interface. It would be much nicer to be able to "see" the
393 list of files backed up in typical GUI tree format.
394 The same mechanism could also be used for creating
395 ad-hoc backup FileSets (item 8).
397 Why: Ease of use -- especially for the end user.
399 Notes: Rather than implementing in Gtk, we probably should go directly
400 for a Browser implementation, even if doing so meant the
401 capability wouldn't be available until much later. Not only
402 is there the question of Windows sites, most
403 Solaris/HP/IRIX, etc, shops can't currently run Gtk programs
404 without installing lots of stuff admins are very wary about.
405 Real sysadmins will always use the command line anyway, and
406 the user who's doing an interactive restore or backup of his
407 own files will in most cases be on a Windows machine running
411 Item 9: Add SSL to daemon communications.
413 What: This provides for secure communications between the daemons.
415 Why: This would allow doing backup across the Internet without
416 privacy concerns (or with much less concern).
418 Notes: The vast majority of near term potential users will be backing up
419 a single site over a LAN and, correctly or not, they probably
420 won't be concerned with security, at least not enough to go to
421 the trouble to set up keys, etc. to screw things down. We suspect
422 that many users genuinely interested in multi-site backup
423 already run some form of VPN software in their internetwork
424 connections, and are willing to delegate security to that layer.
427 Item 10: Define definitive tape format.
430 What: Define that definitive tape format that will not change
431 for the next millennium.
433 Why: Stability, security.
435 Notes: See notes for item 11 below.
438 Item 11: New daemon communication protocol.
440 What: The current daemon to daemon protocol is basically an ASCII
441 printf() and sending the buffer. On the receiving end, the
442 buffer is sscanf()ed to unpack it. The new scheme would
443 be a binary format that allows quick packing and unpacking
444 of any data type with named fields.
446 Why: Using binary packing would be faster. Named fields will permit
447 error checking to ensure that what is sent is what the
448 receiver really wants.
450 Notes: These are internal improvements in the interest of the
451 long-term stability and evolution of the program. On the one
452 hand, the sooner they're done, the less code we have to rip
453 up when the time comes to install them. On the other hand, they
454 don't bring an immediately perceptible benefit to potential
455 users. Item 10 and possibly item 11 should be deferred until Bacula
456 is well established with a growing user community more or
457 less happy with the feature set. At that time, it will make a
458 good "next generation" upgrade in the interest of data
464 ======================================================
466 It is somewhat like a Full save becomes an incremental since
467 the Base job (or jobs) plus other non-base files.
469 - New BaseFile table that contains:
470 JobId, BaseJobId, FileId (from Base).
471 i.e. for each base file that exists but is not saved because
472 it has not changed, the File daemon sends the JobId, BaseId,
473 and FileId back to the Director who creates the DB entry.
474 - To initiate a Base save, the Director sends the FD
475 the FileId, and full filename for each file in the Base.
476 - When the FD finds a Base file, he requests the Director to
477 send him the full File entry (stat packet plus MD5), or
478 conversely, the FD sends it to the Director and the Director
479 says yes or no. This can be quite rapid if the FileId is kept
480 by the FD for each Base Filename.
481 - It is probably better to have the comparison done by the FD
482 despite the fact that the File entry must be sent across the
484 - An alternative would be to send the FD the whole File entry
485 from the start. The disadvantage is that it requires a lot of
486 space. The advantage is that it requires less communications
488 - The Job record must be updated to indicate that one or more
490 - At end of Job, FD returns:
491 1. Count of base files/bytes not written to tape (i.e. matches)
492 2. Count of base file that were saved i.e. had changed.
493 - No tape record would be written for a Base file that matches, in the
494 same way that no tape record is written for Incremental jobs where
495 the file is not saved because it is unchanged.
496 - On a restore, all the Base file records must explicitly be
497 found from the BaseFile tape. I.e. for each Full save that is marked
498 to have one or more Base Jobs, search the BaseFile for all occurrences
500 - An optimization might be to make the BaseFile have:
506 This would avoid the need to explicitly fetch each File record for
507 the Base job. The Base Job record will be fetched to get the
508 VolSessionId and VolSessionTime.
509 =========================================================
511 ==========================================================
513 For each Incremental job that is run, there may be files that
514 were found but not saved because they were locked (this applies
515 only to Windows). Such a system could send back to the Director
516 a list of Unsaved files.
518 - New UnSavedFiles table that contains:
522 - Then in the next Incremental job, the list of Unsaved Files will be
523 feed to the FD, who will ensure that they are explicitly chosen even
524 if standard date/time check would not have selected them.
525 =============================================================
529 =============================================================
531 Request For Comments For File Backup Options
534 Subject: File Backup Options
537 A few days ago, a Bacula user who is backing up to file volumes and
538 using compression asked if it was possible to suppress compressing
539 all .gz files since it was a waste of CPU time. Although Bacula
540 currently permits using different options (compression, ...) on
541 a directory by directory basis, it cannot do it on a file by
542 file basis, which is clearly what was desired.
544 Proposed Implementation:
545 To solve this problem, I propose the following:
547 - Add a new Director resource type called FileOptions.
549 - The FileOptions resource will have records for all
550 options that can currently be specified on the Include record
551 (in a FileSet). Examples below.
553 - The FileOptions resource will permit an exclude option as well
554 as a number of additional options.
556 - The heart of the FileOptions resource is the ability to
557 supply any number of ApplyTo records which specify POSIX
558 regular expressions. These ApplyTo regular expressions are
559 applied to the fully qualified filename (path and all). If
560 one matches, then the FileOptions will be used.
562 - When an ApplyTo specification matches an included file, the
563 options specified in the FileOptions resource will override
564 the default options specified on the Include record.
566 - Include records will be modified to permit referencing one or
567 more FileOptions resources. The FileOptions will be used
568 in the order listed on the Include record and the first
569 one that matches will be applied.
571 - Options (or specifications) currently supplied on the Include
572 record will be deprecated (i.e. removed in a later version a
573 year or so from now).
575 - The Exclude record will be deprecated as the same functionality
576 can be obtained by using an Exclude = yes in the FileOptions.
579 The following records can appear in the FileOptions resource. An
580 asterisk preceding the name indicates a feature not currently
584 - Compression= (GZIP, ...)
585 - Signature= (MD5, SHA1, ...)
587 - OneFs= (yes/no) - remain on one filesystem
588 - Recurse= (yes/no) - recurse into subdirectories
589 - Sparse= (yes/no) - do sparse file backup
590 - *Exclude= (yes/no) - exclude file from being saved
591 - *Reader= (filename) - external read (backup) program
592 - *Plugin= (filename) - read/write plugin module
595 - verify= (ipnougsamc5) - verify options
598 - replace= (always/ifnewer/ifolder/never) - replace options currently
600 - *Writer= (filename) - external write (restore) program
604 Currently options specifying compression, MD5 signatures, recursion,
605 ... of a FileSet are supplied on the Include record. These will now
606 all be collected into a FileOptions resource, which will be
607 specified on the Include in place of the options. Multiple FileOptions
608 may be specified. Since the FileOptions contain regular expressions
609 that are applied to the full filename, this will give the ability
610 to specify backup options on a file by file basis to whatever level
619 Include = compression=GZIP signature=MD5 {
628 Include = FileOptions=Opts {
639 That's a lot more to do the same thing, but it gives the ability to
640 apply options on a file by file basis. For example, suppose you
641 want to compress all files but not any file with extensions .gz or .Z.
642 You could do so as follows:
646 Include = FileOptions=NoCompress FileOptions=Opts {
654 ApplyTo = /*.?*/ # matches all files
659 # Note multiple ApplyTos are ORed
660 ApplyTo = /*.gz/ # matches .gz files */
661 ApplyTo = /*.Z/ # matches .Z files */
664 Now, since the NoCompress FileOptions is specified first on the
665 Include line, any *.gz or *.Z file will have an MD5 signature computed,
666 but will not be compressed. For all other files, the NoCompress will not
667 match, so the Opts options will be used which will include GZIP
671 - Is it necessary to provide some means of ANDing regular expressions
672 and negation? (not currently planned)
674 e.g. ApplyTo = /*.gz/ && !/big.gz/
676 - I see that Networker has a "null" module which, if specified, does not
677 backup the file, but does make an record of the file in the catalog
678 so that the catalog will reflect an exact picture of the filesystem.
679 The result is that the file can be "seen" when "browsing" the save
680 sets, but it cannot be restored.
682 Is this really useful? Should it be implemented in Bacula?
685 After implementing the above, the user will be able to specify
686 on a file by file basis (using regular expressions) what options are
687 applied for the backup.
688 ====================================
690 =========================================
691 Proposal by Bill Sellers
693 Return-Path: <w.a.sellers@larc.nasa.gov>
694 Received: from post.larc.nasa.gov (post.larc.nasa.gov [128.155.4.45]) by matou.sibbald.com (8.11.6/8.11.6) with ESMTP id h0ELUIm07622 for <kern@sibbald.com>; Tue, 14 Jan 2003 22:30:18 +0100
695 Received: from Baron.larc.nasa.gov (baron.larc.nasa.gov [128.155.40.132]) by post.larc.nasa.gov (pohub4.6) with ESMTP id QAA09768 for <kern@sibbald.com>; Tue, 14 Jan 2003 16:30:14 -0500 (EST)
696 Message-Id: <5.1.0.14.2.20030114153452.028dbae8@pop.larc.nasa.gov>
697 X-Sender: w.a.sellers@pop.larc.nasa.gov
698 X-Mailer: QUALCOMM Windows Eudora Version 5.1
699 Date: Tue, 14 Jan 2003 16:30:18 -0500
700 To: Kern Sibbald <kern@sibbald.com>
701 From: Bill Sellers <w.a.sellers@larc.nasa.gov>
702 Subject: Re: [Bacula-users] Bacula remote storage?
703 In-Reply-To: <1042565382.1845.177.camel@rufus>
704 References: <5.1.0.14.2.20030114113004.0293a210@pop.larc.nasa.gov> <5.1.0.14.2.20030113170650.028dad88@pop.larc.nasa.gov> <5.1.0.14.2.20030113170650.028dad88@pop.larc.nasa.gov> <5.1.0.14.2.20030114113004.0293a210@pop.larc.nasa.gov>
706 Content-Type: text/plain; charset="us-ascii"; format=flowed
707 X-Annoyance-Filter-Junk-Probability: 0
708 X-Annoyance-Filter-Classification: Mail
709 At 06:29 PM 1/14/2003 +0100, you wrote:
712 >Well, if you cannot put a Bacula client on the machine,
713 >then it is a big problem. If you know of some software
714 >that can do what you want, let me know, because I
715 >really just don't know how to do it -- at least not
721 We have been able to get Amanda to use the HSM as a storage
722 device. Someone here wrote a driver for Amanda. BUT, Amanda doesn't
723 handle Windows systems very well (or at all without Samba). So I am
724 looking for a backup system that has a Windows client. I really like the
725 Windows integration of Bacula.
727 From the command line, its rather trivial to move the data around. We use
730 tar cf - ./files | gzip -c | rsh hsm dd of=path/file.tgz
732 or if you use GNU tar:
734 tar czf hsm:path/file.tgz ./files
736 One idea for you to consider; Sendmail offers pipes in the aliases file;
737 (mailpipe: "|/usr/bin/vacation root") and Perl supports pipes in the
738 "open" statement (open FILE, "|/bin/nroff -man";) Could you could make a
739 pipe available, as a storage device? Then we could use any command that
740 handles stdin as a storage destination.
747 #Password is not used in rsh, but might be used in ftp.
748 Device = "| gzip -c | rsh hsm dd of=path/file.tgz"
755 Password = "foobar&-"
756 Device = "| ncftpput -c hsm /path/file.bacula"
760 >If you have some local storage available, you could
761 >use Bacula to backup to disk volumes, then use some
762 >other software (ftp, scp) to move them to the HSM
763 >machine. However, this is a bit kludgy.
766 It is, but maybe worth a try. Is there some function in Bacula to put
767 variables in filenames? i.e. backup.2003-01-15.root
774 w.a.sellers@larc.nasa.gov
776 ==============================================
777 The Project for the above
779 I finally realized that this is not at all
780 the same as reader/writer programs or plugins,
781 which are alternate ways of accessing the
782 files to be backed up. Rather, it is an alternate
783 form of storage device, and I have always planned
784 that Bacula should be able to handle all sorts
787 So, I propose the following phases:
789 1. OK from you to invest some time in testing
790 this as I implement it (requires that you
791 know how to download from the SourceForge
792 cvs -- which I imagine is a piece of cake
795 2. Dumb implementation by allowing a device to
796 be a fifo for write only.
797 Reason: easy to implement, proof of concept.
799 3. Try reading from fifo but with fixed block
801 Reason: proof of concept, easy to implement.
803 4. Extend reading from fifo (restores) to handle
805 Reason: requires some delicate low level coding
806 which could destabilize all of Bacula.
808 5. Implementation of above but to a program. E.g.
809 Device = "|program" (not full pipeline).
810 Reason: routines already exist, and program can
811 be a shell script which contains anything.
813 6. Add full pipeline as a possibility. E.g.
814 Device = "| gzip -c | rsh hsm dd of=path/file.tgz"
815 Reason: needs additional coding to implement full
816 pipeline (must fire off either a shell or all
817 programs and connect their pipes).
819 There are a good number of details in each step
820 that I have left out, but I will specify them at
821 every stage, and there may be a few changes as things
822 evolve. I expect that to get to stage 5 will take a
823 few weeks, and at that point, you will have
824 everything you need (just inside a script).
825 Stage 6 will probably take longer, but if this
826 project pleases you, what we do for 5 should
827 be adequate for some time.
831 =============================================
833 Done: (see kernsdone for more)
834 - Look into Pruning/purging problems or why there seem to
835 be so many files listed each night.
836 - Fix cancel in find_one -- need jcr.
837 - Cancel does not work for restore in FD.
838 - Write SetJobStatus() function so cancel status not lost.
839 - Add include list to end of chain in findlib
840 - Zap sd_auth_key after use
841 - Add Bar code reading capabilities (new mtx-changer)
842 - Figure out some way to automatically backup all local partitions
843 - Make hash table for linked files in findlib/find_one.c:161
845 - Rewrite find_one.c to use only pool_memory instead of
846 alloca and malloc (probably not necessary).
847 - Make sure btraceback goes into /sbin not sysconf directory.
848 - InitVerify is getting pruned and it shouldn't (document it)
849 - Make 1.28c release ??? NO do 1.29 directly
850 - Set timeout on opening fifo for save or restore (findlib)
851 - Document FIFO storage device.
852 - Document fifo and | and <
855 - Get correct error status from run_program or open_bpipe().
856 - Restrict permissions on File Volumes (now 0640).
857 - Umasked 022 daemons
858 - Fix restore of hard linked file.
859 - Figure out how to allow multiple simultaneous file Volumes on a single device.
860 - Implement multiple simultaneous file Volumes on a single device.
861 - Cleanup db_update_media and db_update_pool
862 - Flush all the daemon messages at the end of every job.
863 - Change stat1= fgets()!=NULL to stat1=fgest()==NULL; in
864 run_program -- bpipe.c
865 - Apparently cancel does not work for jobs waiting to be
867 - Implement TCP/IP connection for MySQL
868 - Pull a canceled job from the Scheduling queue.
869 - Implement max_file_size in block.c (already done, just tweaked).
870 - Look at purge jobs volume (at least document it, and see if it is
872 - Add list volumes does all pools. list volumes pool=xxx now works.
873 - Add pool= to "list media" in ua_output.c
874 - Strip trailing slashes from Include directory names in the FD.
875 - Fix Error: bnet.c:408 gethostbyname() for lpmatou failed:
876 ERR=Operation not permited loop.
877 - Add code if there is no mtio.h (cannot do -- too many ioctl defines needed)
878 - Produce better error messages in when error/eof writing block.
879 - Cancelling of a queued job does NOT work!!!!!!
881 rufus-dir: Volume used once. Marking Volume "File0003" as Used.
882 rufus-sd: Recycled volume File0003 on device /home/kern/bacula/working, all previous data lost.
883 rufus-dir: Volume used once. Marking Volume "File0003" as Used.
884 - Ability to backup to a file then later transfer to a tape -- Migration.
885 Migration based on MaxJobs(MinJobs),MaxVols(MinVols),AgeJobs,MaxBytes(MinBytes)
886 (i.e. HighwaterMark, LowwaterMark).
887 - Eugeny Fisher <efischer@vip-rus.com> wants to cycle through a
888 set of volumes recycling the oldest volume when it is needed.
889 - gethostbyname failure in bnet_connect() continues
890 generating errors -- should stop.
891 - Add chflags() code for FreeBSD file flags
892 - Bevan Anderson suggests having a run queue for each device
893 so that multiple simultaneous jobs can run but each writing
894 to a different Volume.
895 - Look at handling <> in smtp doesn't work with exim.
896 - Need to specify MaximumConcurrentJobs in the Job resource.
897 - ***test GetFileAttributexEx, and remove MessageBox at 335 of winservice.cpp ****
898 - Implement finer multiprocessing options.
899 - Implement | and < in Exclude statements.
900 - Figure out some way to specify a retention period for files
901 that no longer exist on the machine -- so that we maintain
902 say backups for 30 days, but if the file is deleted, we maintain
903 the last copy for 1 year. -- answer Volume retention.
904 - Make non-zero status from RunJobBefore/After error the job.
905 - Need define of int_least16_t in sha1.h for SuSE.
906 - Implement bar code reader for autochangers
907 - Document new MaximumConcurrentJob records (Job, Client, Storage)
908 - Write up how to use/manage disk Volume Storage. ******
909 - Remove kern and kelvin from mysql_grant...
910 - Install grant_mysql...
911 - Strip trailing / from Include
912 - add #define ENABLE_NLS for Gnome compile on SuSE.
913 - Add Client FS/OS id (Linux, Win95/98, ...).
914 - Concept of VolumeSet during restore which is a list
915 of Volume names needed.
916 - Turn virutally all sprintfs into snprintfs.
917 - Update volume=Test01 requests pool, then lists volumes.
918 **** Test select_pool_and_media ...
920 - Add IP address to authentication failures.
921 - Add a default File storage so that new users can do backup
922 and restores right away.