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 fifo and | and <
9 - Document how to automatically backup all local partitions
10 - Document problems with Verify and pruning.
11 - Document how to use multiple databases.
12 - Document FIFO storage device.
15 Testing to do: (painful)
16 - that console command line options work
17 - blocksize recognition code.
20 - Add whereabslinks to where or not where absolute links
23 - Get correct error status from run_program or open_bpipe().
24 - Look at handling <> in smtp doesn't work with exim.
25 - Priority job to go to top of list.
26 - Implement Bar code handling
27 - Why is catreq.c:111 Find vol called twice for a job?
28 - Find out why Full saves run slower and slower (hashing?)
29 - Figure out how to allow multiple simultaneous file Volumes on a single device.
30 - Why are save/restore of device different sizes (sparse?) Yup! Fix it.
31 - Implement some way for the Console to dynamically create a job.
32 - Restore to a particular time -- e.g. before date, after date.
34 - Implement disk spooling
35 - Implement finer multiprocessing options.
36 - Solaris -I on tar for include list
37 - Enable avoid backing up archive device (findlib/find_one.c:128)
38 - Implement FileOptions (see end of this document)
39 - Implement Bacula plugins -- design API
40 - Make bcopy read through bad tape records.
41 - Need a verbose mode in restore, perhaps to bsr.
42 - bscan without -v is too quiet -- perhaps show jobs.
43 - Add code to reject whole blocks if not wanted on restore.
44 - Implement multiple simultaneous file Volumes on a single device.
45 - Start working on Base jobs.
46 - Make sure the MaxVolFiles is fully implemented in SD
47 - Flush all the daemon messages at the end of every job.
48 - Check if both CatalogFiles and UseCatalog are set to SD.
49 - Check if we can increase Bacula FD priorty in Win2000
50 - Need return status on read_cb() from read_records(). Need multiple
51 records -- one per Job, maybe a JCR or some other structure with
53 - Figure out how to do a bare metal Windows restore
54 - Fix read_record to handle multiple sessions.
55 - Program files (i.e. execute a program to read/write files).
56 Pass read date of last backup, size of file last time.
57 - Put system type returned by FD into catalog.
58 - Possibly add email to Watchdog if drive is unmounted too
59 long and a job is waiting on the drive.
60 - Strip trailing slashes from Include directory names in the FD.
61 - Use read_record.c in SD code.
62 - Why don't we get an error message from Win32 FD when bootstrap
63 file cannot be created for restore command?
64 - Need to specify MaximumConcurrentJobs in the Job resource.
65 - When Marking a file in Restore that is a hard link, also
66 mark the link so that the data will be reloaded.
67 - Restore program that errors in SD due to no tape reports
68 OK incorrectly in output.
69 - After unmount, if restore job started, ask to mount.
70 - Make Restore report an error if FD or SD term codes are not OK.
71 - Convert all %x substitution variables, which are hard to remember
72 and read to %(variable-name). Idea from TMDA.
73 - Add JobLevel in FD status (but make sure it is defined).
74 - Make Pool resource handle Counter resources.
75 - Remove NextId for SQLite. Optimize.
76 - Strip trailing / from Include
77 - Move all SQL statements into a single location.
78 - Cleanup db_update_media and db_update_pool
79 - Add UA rc and history files.
80 - put termcap (used by console) in ./configure and
81 allow -with-termcap-dir.
82 - Enhance time and size scanning routines.
83 - Fix Autoprune for Volumes to respect need for full save.
84 - Fix Win32 config file definition name on /install
85 - No READLINE_SRC if found in alternate directory.
86 - Add Client FS/OS id (Linux, Win95/98, ...).
87 - Test a second language e.g. french.
88 - Compare tape to Client files (attributes, or attributes and data)
89 - Make all database Ids 64 bit.
90 - Write an applet for Linux.
91 - Add estimate to Console commands
92 - Find solution to blank filename (i.e. path only) problem.
93 - Implement new daemon communications protocol.
94 - Remove PoolId from Job table, it exists in Media.
95 - Allow console commands to detach or run in background.
96 - Fix status delay on storage daemon during rewind.
97 - Add SD message variables to control operator wait time
98 - Maximum Operator Wait
99 - Minimum Message Interval
100 - Maximum Message Interval
101 - Send Operator message when cannot read tape label.
102 - Verify level=Volume (scan only), level=Data (compare of data to file).
103 Verify level=Catalog, level=InitCatalog
105 - Add keyword search to show command in Console.
106 - Fix Win2000 error with no messages during startup.
107 - Events : tape has more than xxx bytes.
108 - Restrict characters permitted in a Resource name.
109 - Complete code in Bacula Resources -- this will permit
110 reading a new config file at any time.
111 - Handle ctl-c in Console
112 - Implement LabelTemplate (at least first cut).
113 - Implement script driven addition of File daemon to config files.
114 - Think about how to make Bacula work better with File (non-tape) archives.
115 - Write Unix emulator for Windows.
117 - Implement new serialize subroutines
118 send(socket, "string", &Vol, "uint32", &i, NULL)
119 - Audit all UA commands to ensure that we always prompt where possible.
120 - If ./btape is called without /dev, assume argument is a Storage resource name.
121 - Put memory utilization in Status output of each daemon
122 if full status requested or if some level of debug on.
123 - Make database type selectable by .conf files i.e. at runtime
124 - gethostbyname failure in bnet_connect() continues
125 generating errors -- should stop.
126 - Set flag for uname -a. Add to Volume label.
127 - Implement throttled work queue.
128 - Check for EOT at ENOSPC or EIO or ENXIO (unix Pc)
129 - Allow multiple Storage specifications (or multiple names on
130 a single Storage specification) in the Job record. Thus a job
131 can be backed up to a number of storage devices.
132 - Implement dump label to UA
133 - Concept of VolumeSet during restore which is a list
134 of Volume names needed.
135 - Restore files modified after date
136 - Restore file modified before date
137 - Emergency restore info:
139 - Backup working directory
141 - Restore -- do nothing but show what would happen
142 - SET LD_RUN_PATH=$HOME/mysql/lib/mysql
143 - Implement Restore FileSet=
144 - Create a protocol.h and protocol.c where all protocol messages
146 - If SD cannot open a drive, make it periodically retry.
147 - Remove duplicate fields from jcr (e.g. jcr.level and jcr.jr.Level, ...).
148 - Timout a job or terminate if link goes down, or reopen link and query.
149 - Find general solution for sscanf size problems (as well
150 as sprintf. Do at run time?
151 - Concept of precious tapes (cannot be reused).
152 - Make bcopy copy with a single tape drive.
153 - Permit changing ownership during restore.
155 - Autolabel should be specified by DIR instead of SD.
156 - Find out how to get the system tape block limits, e.g.:
157 Apr 22 21:22:10 polymatou kernel: st1: Block limits 1 - 245760 bytes.
158 Apr 22 21:22:10 polymatou kernel: st0: Block limits 2 - 16777214 bytes.
161 - AutoScan (check checksum of tape)
162 - Format command = "format /dev/nst0"
166 - Seek resolution (usually corresponds to buffer size)
167 - EODErrorCode=ENOSPC or code
168 - Partial Read error code
169 - Partial write error code
170 - Nonformatted read error
171 - Nonformatted write error
172 - WriteProtected error
176 - IgnoreCloseErrors=yes
186 - FD sends unsaved file list to Director at end of job (see
188 - File daemon should build list of files skipped, and then
189 at end of save retry and report any errors.
190 - Write a Storage daemon that uses pipes and
191 standard Unix programs to write to the tape.
193 - Need something that monitors the JCR queue and
194 times out jobs by asking the deamons where they are.
195 - Enhance Jmsg code to permit buffering and saving to disk.
196 - device driver = "xxxx" for drives.
197 - restart: paranoid: read label fsf to
198 eom read append block, and go
199 super-paranoid: read label, read all files
200 in between, read append block, and go
201 verify: backspace, read append block, and go
202 permissive: same as above but frees drive
203 if tape is not valid.
205 - Ensure that /dev/null works
206 - Need report class for messages. Perhaps
207 report resource where report=group of messages
208 - enhance scan_attrib and rename scan_jobtype, and
209 fill in code for "since" option
210 - Director needs a time after which the report status is sent
211 anyway -- or better yet, a retry time for the job.
212 Don't reschedule a job if previous incarnation is still running.
213 - Some way to automatically backup everything is needed????
214 - Need a structure for pending actions:
216 - termination status (part of buffered msgs?)
217 - Concept of grouping Storage devices and job can use
218 any of a number of devices
220 Read, Write, Clean, Delete
221 - Login to Bacula; Bacula users with different permissions:
222 owner, group, user, quotas
223 - Store info on each file system type (probably in the job header on tape.
224 This could be the output of df; or perhaps some sort of /etc/mtab record.
227 - Design at hierarchial storage for Bacula.
228 - Implement FSM (File System Modules).
229 - Identify unchanged or "system" files and save them to a
230 special tape thus removing them from the standard
231 backup FileSet -- BASE backup.
232 - Turn virutally all sprintfs into snprintfs.
233 - Heartbeat between daemons.
234 - Audit M_ error codes to ensure they are correct and consistent.
235 - Add variable break characters to lex analyzer.
236 Either a bit mask or a string of chars so that
237 the caller can change the break characters.
238 - Make a single T_BREAK to replace T_COMMA, etc.
239 - Ensure that File daemon and Storage daemon can
240 continue a save if the Director goes down (this
241 is NOT currently the case). Must detect socket error,
242 buffer messages for later.
243 - Enhance time/duration input to allow multiple qualifiers e.g. 3d2h
244 - Add ability to backup to two Storage devices (two SD sessions) at
245 the same time -- e.g. onsite, offsite.
246 - Add the ability to consolidate old backup sets (basically do a restore
247 to tape and appropriately update the catalog). Compress Volume sets.
248 Might need to spool via file is only one drive is available.
249 - Compress or consolidate Volumes of old possibly deleted files. Perhaps
250 someway to do so with every volume that has less than x% valid
255 Bacula Projects Roadmap
257 last update 5 January 2003
259 Item 1: Multiple simultaneous Jobs. (done)
260 Done -- Restore part needs better implementation to work correctly
262 What: Permit multiple simultaneous jobs in Bacula.
264 Why: An enterprise level solution needs to go fast without the
265 need for the system administrator to carefully tweak
266 timing. Based on the benchmarks, during a full
267 backup, NetWorker typically hit 10 times the bandwidth to
268 the tape compared to Bacula--largely. This is probably due to
269 running parallel jobs and multi-threaded filling of buffers
270 and writing them to tape. This should also make things work
271 better when you have a mix of fast and slow machines backing
274 Notes: Bacula was designed to run multiple simultaneous jobs. Thus
275 implementing this is a matter of some small cleanups and
279 Item 2: Make the Storage daemon use intermediate file storage to buffer data.
280 Deferred -- not necessary yet.
282 What: If data is coming into the SD too fast, buffer it to
283 disk if the user has configured this option.
285 Why: This would be nice, especially if it more or less falls out
286 when implementing (1) above. If not, it probably should not
287 be given a high priority because fundamentally the backup time
288 is limited by the tape bandwidth. Even though you may finish a
289 client job quicker by spilling to disk, you still have to
290 eventually get it onto tape. If intermediate disk buffering
291 allows us to improve write bandwidth to tape, it may make
294 Notes: Whether or not this is implemented will depend upon performance
295 testing after item 1 is implemented.
298 Item 3: Write the bscan program -- also write a bcopy program.
301 What: Write a program that reads a Bacula tape and puts all the
302 appropriate data into the catalog. This allows recovery
303 from a tape that is no longer in the database, or it allows
304 re-creation of a database if lost.
306 Why: This is a fundamental robustness and disaster recovery tool
307 which will increase the comfort level of a sysadmin
308 considering adopting Bacula.
310 Notes: A skeleton of this program already exists, but much work
311 needs to be done. Implementing this will also make apparent
312 any deficiencies in the current Bacula tape format.
315 Item 4: Implement Base jobs.
317 What: A base job is sort of like a Full save except that you
318 will want the FileSet to contain only files that are unlikely
319 to change in the future (i.e. a snapshot of most of your
320 system after installing it). After the base job has been run,
321 when you are doing a Full save, you can specify to exclude
322 all files saved by the base job that have not been modified.
324 Why: This is something none of the competition does, as far as we know
325 (except BackupPC, which is a Perl program that saves to disk
326 only). It is big win for the user, it makes Bacula stand out
327 as offering a unique optimization that immediately saves time
330 Notes: Big savings in tape usage. Will require more resources because
331 the e. DIR must send FD a list of files/attribs, and the FD must
332 search the list and compare it for each file to be saved.
335 Item 5: Implement Label templates
337 What: This is a mechanism whereby Bacula can automatically create
338 a tape label for new tapes according to a detailed specification
339 provided by the user.
341 Why: It is a major convenience item for folks who use automated label
344 Notes: Bacula already has a working form of automatic tape label
345 creation, but it is very crude. The design for the complete
346 tape labeling project is already documented in the manual.
349 Item 6: Write a regression script.
352 What: This is an automatic script that runs and tests as many features
353 of Bacula as possible. The output is compared to previous
354 versions of Bacula and any differences are reported.
356 Why: This is an enormous help in preventing introduction of new
357 errors in parts of the program that already work correctly.
359 Notes: This probably should be ranked higher, it's something the typical
360 user doesn't see. Depending on how it's implemented, it may
361 make sense to defer it until the archival tape format and
362 user interface mature.
365 Item 7: GUI for interactive restore
366 Item 8: GUI for interactive backup
368 What: The current interactive restore is implemented with a tty
369 interface. It would be much nicer to be able to "see" the
370 list of files backed up in typical GUI tree format.
371 The same mechanism could also be used for creating
372 ad-hoc backup FileSets (item 8).
374 Why: Ease of use -- especially for the end user.
376 Notes: Rather than implementing in Gtk, we probably should go directly
377 for a Browser implementation, even if doing so meant the
378 capability wouldn't be available until much later. Not only
379 is there the question of Windows sites, most
380 Solaris/HP/IRIX, etc, shops can't currently run Gtk programs
381 without installing lots of stuff admins are very wary about.
382 Real sysadmins will always use the command line anyway, and
383 the user who's doing an interactive restore or backup of his
384 own files will in most cases be on a Windows machine running
388 Item 9: Add SSL to daemon communications.
390 What: This provides for secure communications between the daemons.
392 Why: This would allow doing backup across the Internet without
393 privacy concerns (or with much less concern).
395 Notes: The vast majority of near term potential users will be backing up
396 a single site over a LAN and, correctly or not, they probably
397 won't be concerned with security, at least not enough to go to
398 the trouble to set up keys, etc. to screw things down. We suspect
399 that many users genuinely interested in multi-site backup
400 already run some form of VPN software in their internetwork
401 connections, and are willing to delegate security to that layer.
404 Item 10: Define definitive tape format.
407 What: Define that definitive tape format that will not change
408 for the next millennium.
410 Why: Stability, security.
412 Notes: See notes for item 11 below.
415 Item 11: New daemon communication protocol.
417 What: The current daemon to daemon protocol is basically an ASCII
418 printf() and sending the buffer. On the receiving end, the
419 buffer is sscanf()ed to unpack it. The new scheme would
420 be a binary format that allows quick packing and unpacking
421 of any data type with named fields.
423 Why: Using binary packing would be faster. Named fields will permit
424 error checking to ensure that what is sent is what the
425 receiver really wants.
427 Notes: These are internal improvements in the interest of the
428 long-term stability and evolution of the program. On the one
429 hand, the sooner they're done, the less code we have to rip
430 up when the time comes to install them. On the other hand, they
431 don't bring an immediately perceptible benefit to potential
432 users. Item 10 and possibly item 11 should be deferred until Bacula
433 is well established with a growing user community more or
434 less happy with the feature set. At that time, it will make a
435 good "next generation" upgrade in the interest of data
441 ======================================================
443 It is somewhat like a Full save becomes an incremental since
444 the Base job (or jobs) plus other non-base files.
446 - New BaseFile table that contains:
447 JobId, BaseJobId, FileId (from Base).
448 i.e. for each base file that exists but is not saved because
449 it has not changed, the File daemon sends the JobId, BaseId,
450 and FileId back to the Director who creates the DB entry.
451 - To initiate a Base save, the Director sends the FD
452 the FileId, and full filename for each file in the Base.
453 - When the FD finds a Base file, he requests the Director to
454 send him the full File entry (stat packet plus MD5), or
455 conversely, the FD sends it to the Director and the Director
456 says yes or no. This can be quite rapid if the FileId is kept
457 by the FD for each Base Filename.
458 - It is probably better to have the comparison done by the FD
459 despite the fact that the File entry must be sent across the
461 - An alternative would be to send the FD the whole File entry
462 from the start. The disadvantage is that it requires a lot of
463 space. The advantage is that it requires less communications
465 - The Job record must be updated to indicate that one or more
467 - At end of Job, FD returns:
468 1. Count of base files/bytes not written to tape (i.e. matches)
469 2. Count of base file that were saved i.e. had changed.
470 - No tape record would be written for a Base file that matches, in the
471 same way that no tape record is written for Incremental jobs where
472 the file is not saved because it is unchanged.
473 - On a restore, all the Base file records must explicitly be
474 found from the BaseFile tape. I.e. for each Full save that is marked
475 to have one or more Base Jobs, search the BaseFile for all occurrences
477 - An optimization might be to make the BaseFile have:
483 This would avoid the need to explicitly fetch each File record for
484 the Base job. The Base Job record will be fetched to get the
485 VolSessionId and VolSessionTime.
486 =========================================================
488 ==========================================================
490 For each Incremental job that is run, there may be files that
491 were found but not saved because they were locked (this applies
492 only to Windows). Such a system could send back to the Director
493 a list of Unsaved files.
495 - New UnSavedFiles table that contains:
499 - Then in the next Incremental job, the list of Unsaved Files will be
500 feed to the FD, who will ensure that they are explicitly chosen even
501 if standard date/time check would not have selected them.
502 =============================================================
506 =============================================================
508 Request For Comments For File Backup Options
511 Subject: File Backup Options
514 A few days ago, a Bacula user who is backing up to file volumes and
515 using compression asked if it was possible to suppress compressing
516 all .gz files since it was a waste of CPU time. Although Bacula
517 currently permits using different options (compression, ...) on
518 a directory by directory basis, it cannot do it on a file by
519 file basis, which is clearly what was desired.
521 Proposed Implementation:
522 To solve this problem, I propose the following:
524 - Add a new Director resource type called FileOptions.
526 - The FileOptions resource will have records for all
527 options that can currently be specified on the Include record
528 (in a FileSet). Examples below.
530 - The FileOptions resource will permit an exclude option as well
531 as a number of additional options.
533 - The heart of the FileOptions resource is the ability to
534 supply any number of ApplyTo records which specify POSIX
535 regular expressions. These ApplyTo regular expressions are
536 applied to the fully qualified filename (path and all). If
537 one matches, then the FileOptions will be used.
539 - When an ApplyTo specification matches an included file, the
540 options specified in the FileOptions resource will override
541 the default options specified on the Include record.
543 - Include records will be modified to permit referencing one or
544 more FileOptions resources. The FileOptions will be used
545 in the order listed on the Include record and the first
546 one that matches will be applied.
548 - Options (or specifications) currently supplied on the Include
549 record will be deprecated (i.e. removed in a later version a
550 year or so from now).
552 - The Exclude record will be deprecated as the same functionality
553 can be obtained by using an Exclude = yes in the FileOptions.
556 The following records can appear in the FileOptions resource. An
557 asterisk preceding the name indicates a feature not currently
561 - Compression= (GZIP, ...)
562 - Signature= (MD5, SHA1, ...)
564 - OneFs= (yes/no) - remain on one filesystem
565 - Recurse= (yes/no) - recurse into subdirectories
566 - Sparse= (yes/no) - do sparse file backup
567 - *Exclude= (yes/no) - exclude file from being saved
568 - *Reader= (filename) - external read (backup) program
569 - *Plugin= (filename) - read/write plugin module
572 - verify= (ipnougsamc5) - verify options
575 - replace= (always/ifnewer/ifolder/never) - replace options currently
577 - *Writer= (filename) - external write (restore) program
581 Currently options specifying compression, MD5 signatures, recursion,
582 ... of a FileSet are supplied on the Include record. These will now
583 all be collected into a FileOptions resource, which will be
584 specified on the Include in place of the options. Multiple FileOptions
585 may be specified. Since the FileOptions contain regular expressions
586 that are applied to the full filename, this will give the ability
587 to specify backup options on a file by file basis to whatever level
596 Include = compression=GZIP signature=MD5 {
605 Include = FileOptions=Opts {
616 That's a lot more to do the same thing, but it gives the ability to
617 apply options on a file by file basis. For example, suppose you
618 want to compress all files but not any file with extensions .gz or .Z.
619 You could do so as follows:
623 Include = FileOptions=NoCompress FileOptions=Opts {
631 ApplyTo = /*.?*/ # matches all files
636 # Note multiple ApplyTos are ORed
637 ApplyTo = /*.gz/ # matches .gz files */
638 ApplyTo = /*.Z/ # matches .Z files */
641 Now, since the NoCompress FileOptions is specified first on the
642 Include line, any *.gz or *.Z file will have an MD5 signature computed,
643 but will not be compressed. For all other files, the NoCompress will not
644 match, so the Opts options will be used which will include GZIP
648 - Is it necessary to provide some means of ANDing regular expressions
649 and negation? (not currently planned)
651 e.g. ApplyTo = /*.gz/ && !/big.gz/
653 - I see that Networker has a "null" module which, if specified, does not
654 backup the file, but does make an record of the file in the catalog
655 so that the catalog will reflect an exact picture of the filesystem.
656 The result is that the file can be "seen" when "browsing" the save
657 sets, but it cannot be restored.
659 Is this really useful? Should it be implemented in Bacula?
662 After implementing the above, the user will be able to specify
663 on a file by file basis (using regular expressions) what options are
664 applied for the backup.
665 ====================================
667 =========================================
668 Proposal by Bill Sellers
670 Return-Path: <w.a.sellers@larc.nasa.gov>
671 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
672 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)
673 Message-Id: <5.1.0.14.2.20030114153452.028dbae8@pop.larc.nasa.gov>
674 X-Sender: w.a.sellers@pop.larc.nasa.gov
675 X-Mailer: QUALCOMM Windows Eudora Version 5.1
676 Date: Tue, 14 Jan 2003 16:30:18 -0500
677 To: Kern Sibbald <kern@sibbald.com>
678 From: Bill Sellers <w.a.sellers@larc.nasa.gov>
679 Subject: Re: [Bacula-users] Bacula remote storage?
680 In-Reply-To: <1042565382.1845.177.camel@rufus>
681 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>
683 Content-Type: text/plain; charset="us-ascii"; format=flowed
684 X-Annoyance-Filter-Junk-Probability: 0
685 X-Annoyance-Filter-Classification: Mail
686 At 06:29 PM 1/14/2003 +0100, you wrote:
689 >Well, if you cannot put a Bacula client on the machine,
690 >then it is a big problem. If you know of some software
691 >that can do what you want, let me know, because I
692 >really just don't know how to do it -- at least not
698 We have been able to get Amanda to use the HSM as a storage
699 device. Someone here wrote a driver for Amanda. BUT, Amanda doesn't
700 handle Windows systems very well (or at all without Samba). So I am
701 looking for a backup system that has a Windows client. I really like the
702 Windows integration of Bacula.
704 From the command line, its rather trivial to move the data around. We use
707 tar cf - ./files | gzip -c | rsh hsm dd of=path/file.tgz
709 or if you use GNU tar:
711 tar czf hsm:path/file.tgz ./files
713 One idea for you to consider; Sendmail offers pipes in the aliases file;
714 (mailpipe: "|/usr/bin/vacation root") and Perl supports pipes in the
715 "open" statement (open FILE, "|/bin/nroff -man";) Could you could make a
716 pipe available, as a storage device? Then we could use any command that
717 handles stdin as a storage destination.
724 #Password is not used in rsh, but might be used in ftp.
725 Device = "| gzip -c | rsh hsm dd of=path/file.tgz"
732 Password = "foobar&-"
733 Device = "| ncftpput -c hsm /path/file.bacula"
737 >If you have some local storage available, you could
738 >use Bacula to backup to disk volumes, then use some
739 >other software (ftp, scp) to move them to the HSM
740 >machine. However, this is a bit kludgy.
743 It is, but maybe worth a try. Is there some function in Bacula to put
744 variables in filenames? i.e. backup.2003-01-15.root
751 w.a.sellers@larc.nasa.gov
753 ==============================================
754 The Project for the above
756 I finally realized that this is not at all
757 the same as reader/writer programs or plugins,
758 which are alternate ways of accessing the
759 files to be backed up. Rather, it is an alternate
760 form of storage device, and I have always planned
761 that Bacula should be able to handle all sorts
764 So, I propose the following phases:
766 1. OK from you to invest some time in testing
767 this as I implement it (requires that you
768 know how to download from the SourceForge
769 cvs -- which I imagine is a piece of cake
772 2. Dumb implementation by allowing a device to
773 be a fifo for write only.
774 Reason: easy to implement, proof of concept.
776 3. Try reading from fifo but with fixed block
778 Reason: proof of concept, easy to implement.
780 4. Extend reading from fifo (restores) to handle
782 Reason: requires some delicate low level coding
783 which could destabilize all of Bacula.
785 5. Implementation of above but to a program. E.g.
786 Device = "|program" (not full pipeline).
787 Reason: routines already exist, and program can
788 be a shell script which contains anything.
790 6. Add full pipeline as a possibility. E.g.
791 Device = "| gzip -c | rsh hsm dd of=path/file.tgz"
792 Reason: needs additional coding to implement full
793 pipeline (must fire off either a shell or all
794 programs and connect their pipes).
796 There are a good number of details in each step
797 that I have left out, but I will specify them at
798 every stage, and there may be a few changes as things
799 evolve. I expect that to get to stage 5 will take a
800 few weeks, and at that point, you will have
801 everything you need (just inside a script).
802 Stage 6 will probably take longer, but if this
803 project pleases you, what we do for 5 should
804 be adequate for some time.
808 =============================================
810 Done: (see kernsdone for more)
811 - Look into Pruning/purging problems or why there seem to
812 be so many files listed each night.
813 - Fix cancel in find_one -- need jcr.
814 - Cancel does not work for restore in FD.
815 - Write SetJobStatus() function so cancel status not lost.
816 - Add include list to end of chain in findlib
817 - Zap sd_auth_key after use
818 - Add Bar code reading capabilities (new mtx-changer)
819 - Figure out some way to automatically backup all local partitions
820 - Make hash table for linked files in findlib/find_one.c:161
822 - Rewrite find_one.c to use only pool_memory instead of
823 alloca and malloc (probably not necessary).
824 - Make sure btraceback goes into /sbin not sysconf directory.
825 - InitVerify is getting pruned and it shouldn't (document it)
826 - Make 1.28c release ??? NO do 1.29 directly
827 - Set timeout on opening fifo for save or restore (findlib)