[dump.git] / CHANGES
1 $Id: CHANGES,v 1.297 2009/06/18 09:36:52 stelian Exp $
3 Changes between versions 0.4b41 and 0.4b42 (released xxxxxxxxxxxxxxx)
4 =====================================================================
6 1. Fix printout of the 'Connection to' message in dump/restore.
7 Thanks to Dale Wiles <dwiles@users.sourceforge.net> for the
8 bug report and its fix.
10 2. Fix dump -w/-W output for filesystems having the last backup
11 level equal to 0. Thanks to Pascal Bouchareine
12 <kalou@users.sourceforge.net> for reporting the bug.
14 3. Enable restore to handle restoring onto a different SELinux
15 policy from the dump. Thanks to Tony Nelson
16 <tony_nelson@users.sourceforge.net> for the patch.
17 (see http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=189845
18 for details)
20 4. Fix EA set failures when restoring immutable files. Thanks
21 to Andrew Kroeger for the patch.
23 5. Fix "mode file too large" errors when restoring huge backups.
24 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
25 report.
27 6. Add the ability to link against libtinfo library if present,
28 instead of libtermcap or libncurses libraries. Thanks to
29 Adam Tkac <atkac@redhat.com> for the patch.
31 7. When comparing a backup, do not consider to be an error the fact
32 that we encounter files created while dump was in progress.
33 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
34 report.
36 8. Force '-a' to be the default value when running 'restore -C'.
37 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
38 report.
40 9. Many improvements to the 'cron_dump_to_disk' example.
41 Thanks to Aaron S. Hawley <ashawley@users.sourceforge.net> for
42 the patch.
44 10. Fix the QFA generation when extended attributes are backuped
45 along with the inodes. Prior to this fix, the entries for
46 some inodes may get corrupted (duplicated entries or incorrect
47 ones), making the QFA file unusable for repositionning in
48 restore. Many thanks to Kenneth Porter <shiva@sewingwitch.com>
49 for the bug report and the testing of the fix.
51 11. Fix EA restoration in debug mode (Sourceforge bug #1986216).
53 12. Stefan Auracher <star7@onlinehome.de> noticed that the Tower
54 of Hanoi backup strategy documented in the man page wasn't
55 actually based on the Tower of Hanoi algorithm. The man page
56 was updated accordingly.
58 13. Fix missing level information from dump output when using
59 the default level (RedHat bug #493635). Thanks to Adam Tkac
60 <atkac@redhat.com> for the patch.
62 14. Fix the dump man page to reflect the fact that the default
63 dump level is 0 not 9 (RedHat bug #356121). Thanks to Adam Tkac
64 <atkac@redhat.com> for the patch.
66 Changes between versions 0.4b40 and 0.4b41 (released January 2, 2006)
67 =====================================================================
69 1. Fix restore of dumped Access Control Lists. The previous
70 code for EA works fine, but ACLs needed conversion from the
71 ext2/3 disk format to posix_acl format before restoring.
73 2. Fix some issues with restoration of EA on big endian
74 platforms.
76 3. Fix restore when the symtab is over 2GB in size.
78 4. Made the directory hash indexing an optional feature,
79 accessible by the '-H' option of restore, and disable it
80 by default.
82 5. Fixed dump to not include extended attributes information
83 in the toc (archive) file which confused restore -t.
85 Changes between versions 0.4b39 and 0.4b40 (released May 2, 2005)
86 =================================================================
88 1. Changed restore to emit warnings (instead of emitting a fatal
89 error) if a file (or a directory) is unavailable for a
90 comparision (if the user doesn't have the necessary permissions
91 to access it for example). Thanks to Kenneth Porter
92 <shiva@sewingwitch.com> for the bug report.
94 2. Re-done the 'do not save directory entries to non-dumped inodes
95 (excluded from dump)' feature. The previous implementation
96 worked well for excluded directories but not for regular files.
97 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
98 report.
100 3. Fixed a bug in dump where the tape size was miscalculated when
101 the user used -d/-s to specify the tape characteristics. Thanks
102 to Philip Goisman <goisman@physics.arizona.edu> for reporting
103 the bug.
105 4. Fixed another bug introduced in restore with the hashtree
106 implementation. This one caused restore to stop saying
107 "removenode: non-empty directory" in some cases.
109 5. Added support for dumping and restoring ext2/3 extended
110 attributes (EA), like the access control lists (ACL) or
111 the security labels used by SELinux.
113 Changes between versions 0.4b38 and 0.4b39 (released January 21, 2005)
114 ======================================================================
116 1. The newly added dump_on_cd_3 example was buggy, replace it
117 with an updated version from Andrew Basterfield
118 <bob@cemetery.homeunix.org>.
120 2. Made restore to chdir() back into the initial directory when
121 dumping core while aborting a comparision operation. The
122 previous behaviour was to write the corefile at the root of
123 the directory being compared, which could very well be
124 read only and preventing the corefile generation. Thanks
125 to Kenneth Porter <shiva@sewingwitch.com> for the bug report.
127 3. Silenced the failure to call fgetflags() when comparing an
128 entry which has no ext2 attributes (as in lsattr()).
130 4. Fix a brown paper bug in restore -C which broke restore and
131 caused modifications on the filesystem being compared
132 (directories containing a file with the same name as the
133 directory get renamed to RSTTMP...). Thanks to Kenneth Porter
134 <shiva@sewingwitch.com> for finding the bug and helping me
135 reproduce it.
137 5. Made restore -C force the -N flag (no writing allowed on
138 the disk) in order to prevent more bugs like the above one.
140 Changes between versions 0.4b37 and 0.4b38 (released January 7, 2005)
141 =====================================================================
143 1. Fix a couple of troff syntax bugs in the man pages.
144 Thanks to Eric Raymond <esr@thyrsus.com> for the patch.
146 2. Made restore use either libncurses or libtermcap, depending
147 on which one is available at configure time.
149 3. Fixed restore negative size display bug when comparing a
150 dump containing files over 2GB. Thanks to Steve Bonds
151 <sbonds@users.sourceforge.net> for the bug report.
153 4. Do not save directory entries to non-dumped inodes
154 (excluded from dump). This will eliminate the 'missing
155 file' warnings when doing 'restore -C'.
157 5. Fix dump crash when backuping a huge (2TB) filesystem,
158 due to a bogus calculation on the inode map size.
159 Thanks to Kevin B. Haines <K.B.Haines@rl.ac.uk> for
160 submitting the bug and testing the fix.
162 6. Fix a problem in restore where the final \0 in the symbolic
163 link names could have been lost, generating corrupt filenames.
164 Thanks to Kyle Wilson <kyle.wilson@amd.com> for reporting the
165 bug.
167 7. Implemented a hash list for the directory names in restore.
168 The linear list used before caused problems in interactive
169 restores when dealing with directories having thousands of
170 entries. Thanks to Brian Ristuccia <bristuccia@starentnetworks.com>
171 for reporting the bug.
173 8. Improved restore -C, this time including the directory
174 attributes into the comparision.
176 9. Made restore understand tapes containing EA/ACLs (which will
177 be dumped by the next version of dump). In this version
178 extended attributes on the tape are ignored, for full EA/ACL
179 support wait for the next version or try the experimental EA
180 patch.
182 Changes between versions 0.4b36 and 0.4b37 (released July 7, 2004)
183 ==================================================================
185 1. Added the --enable-staticz configure option which enables
186 dump and restore to be linked with static versions of
187 libz and libbz2 (and dynamic versions of all other libraries).
188 This will make Debian users happy, because libz and libbz2
189 were the only needed libraries living in /usr, all the
190 others live in /lib. In case of system emergency, it is
191 better not to have to rely on an extra filesystem.
192 Thanks to Bdale Garbee <bdale@gag.com> for the suggestion.
194 2. Fix compilation on (at least the Linux Debian port to) AMD64.
195 (<ext2fs/ext2_types.h> defines some types (__s64 and __u64)
196 that are also defined by <linux/types.h> (<asm/types.h>) and
197 they conflict).
199 3. Make dump's reading of the dumpdates file a bit more robust,
200 preventing dump from crashing when the dumpdates file has
201 been modified by hand.
203 4. Fixed some offset calculations in dump code which could
204 lead to "bread lseek errors" on large filesystems. Thanks
205 to Bruce Lowekamp <lowekamp@users.sourceforge.net> for
206 reporting this bug and debugging the issue.
208 5. Made dump use the blkid library when searching for devices
209 by label or uuid instead of dump's own routines.
211 6. Corrected a bug in dump where a wrong LABEL=... line in
212 /etc/fstab could prevent dump from dumping unrelated
213 filesystems. Thanks to Bruce Lowekamp
214 <lowekamp@users.sourceforge.net> for reporting the bug.
216 Changes between versions 0.4b35 and 0.4b36 (released April 21, 2004)
217 ====================================================================
219 1. Fixed dump compilation with old gcc versions. Thanks to
220 Mike Castle <dalgoda@users.sourceforge.net> for the patch.
222 2. Fixed some warnings (howmany, roundup, powerof2 redefined)
223 when compiling against a recent glibc version.
225 3. Fixed a bug in restore preventing the read of a dump tape
226 written with Solaris 7 ufsdump. Thanks to Patrick Higgins
227 <phiggins@transzap.com> for reporting the bug and providing
228 the test case.
230 4. Changed dump to enable the creation of volumes of different
231 sizes in a single run (make -B accept a list of values).
232 Patch contributed by Florian Zumbiehl <florz@gmx.de>.
234 5. Use the glibc provided minor() and major() macros instead
235 of our own bitmask implementation. This should be safe for
236 when the major/minor namespace will migrate to 32 bits.
237 Thanks to Zhang Jun <zhangjun@nanjing-fnst.com> for reporting
238 the bug.
240 6. Made explicit in the dump man page that dump will not create
241 a remote file, it will only write to an already existing one.
243 7. Another try at making size estimates better again.
245 8. Put back the inconditional running of the end-of-tape
246 script which was changed in 0.434 to be run only when -M
247 or multiple -f were NOT specified. Some users rely on this
248 feature even when it is combined with -M/-f.
250 9. Fixed restore when restoring huge backups (where rstdir...
251 temporary files are over 2GB). Thanks to
252 Raphael Jaffey <rjaffey@artic.edu> for reporting this,
253 debugging the issue and making the patch.
255 10. Made restore understand FreeBSD UFS2 tapes. Thanks to
256 David <vrtin@users.sourceforge.net> for submitting the bug
257 and providing a test case.
259 11. Made dump work with an arbitrary integer level (no more 10
260 levels only limitation). Thanks to Matthew
261 <msvincen@midway.uchicago.edu> for the patch.
263 Changes between versions 0.4b34 and 0.4b35 (released December 21, 2003)
264 =======================================================================
266 1. Added a note in the dump man page saying that the default
267 blocksize can be 32 if -d is used with a high density argument.
268 Thanks to Antonios Christofides <A.Christofides@itia.ntua.gr>
269 for the patch.
271 2. Fixed configure to correctly understand CPPFLAGS, CFLAGS,
272 LDFLAGS environment variables. Thanks to Arcady Genkin
273 <antipode@users.sourceforge.net> for reporting the bug.
275 3. Made -e/-E options of dump accept an unlimited number of inodes
276 to be excluded, rather than a hardcoded maximum. Thanks to
277 Dietrich Rothe <d-rothe@users.sourceforge.net> for the patch.
279 4. Updated the autoconf system to 2.50. Forced the -D_BSD_SOURCE
280 and -D_USE_BSD_SIGNAL defines in configure in order to solve
281 64bit build problems because quad_t is redifined with a
282 different signature. Thanks to Mike Harris <mharris@redhat.com>
283 for reporting this bug.
285 5. Made restore build on Solaris, making possible to
286 restore Linux's "enhanced" tapes. Thanks to Uwe Gohlke
287 <uwe@ugsoft.de> for the patch.
289 6. Made an extension in the dump tape format capable of saving
290 MacOSX specific inode extensions. Uwe Gohlke <uwe@ugsoft.de>
291 wrote the extension and contributed the restore code back
292 into this codebase. The same extension mechanism will be
293 used in the future to save ACLs...
295 7. Made rmt work correctly with regard to QFA and local/remote
296 files and tapes. The remote access will however work only
297 when the dump provided rmt version is used. If you want to
298 use another rmt server, please do not use the QFA feature.
299 Thanks to Uwe Gohlke <uwe@ugsoft.de> for the patch.
301 Changes between versions 0.4b33 and 0.4b34 (released April 18, 2003)
302 ====================================================================
304 1. Fixed the INSTALL file to reflect the actual install paths.
305 Thanks to David Ronis <ronis@ronispc.chem.mcgill.ca> for
306 reporting the bug.
308 2. Fixed the configure script to only check for headers presence
309 instead of trying to compile them. This should fix issues
310 with old build environments. Thanks to Kari Lempiainen
311 <kari@funky.pp.fi> for reporting the bug.
313 3. Fixed restore to correctly ignore sockets when comparing
314 a dump (as socket cannot be properly restored anyway). Thanks
315 to Gunther Reiszig <gunther@mit.edu> for reporting the bug.
317 4. Fixed restore to correctly access the archive file (-A argument)
318 even when using a remote tape. Thanks to Fabrice Bellet
319 <fabrice@bellet.info> for reporting the bug.
321 5. Fixed (again) handling of long (largefile) seeks in rmt.
322 Thanks to Fabrice Bellet <fabrice@bellet.info> for reporting
323 the bug.
325 6. Fixed restore corner case when dealing with large block sizes
326 dump is able to write now (-b 1024). Thanks to Fabrice Bellet
327 <fabrice@bellet.info> for reporting the bug.
329 7. Fixed a bug preventing dump to access a filesystem having
330 a label exactly 16 bytes in length. Thanks to <minduni@ti-edu.ch>
331 for reporting the bug.
333 8. Made dump store dump dates using explicit timezones, fixing a
334 problem with incremental dumps if the timezone is modified
335 between the dumps. Thanks to Philippe Troin <phil@fifi.org> for
336 the bug report and the patch.
338 9. Fixed a bug encountered when dumping individual files (not full
339 filesystems or directories) and dangling symbolic links happen
340 to be in the list of files. For as far as dump is concerned,
341 dangling symbolic links are allowed, and are dumped as is.
342 Thanks to Jin-su Ahn <jsahn@ee.snu.ac.kr> for reporting the
343 bug and providing the fix.
345 10. Fixed open and creation modes and permissions for QFA and
346 table-of-contents files in dump and restore. Thanks to
347 Philippe Troin <phil@fifi.org> for the patch.
349 11. Fixed the archive file descriptor handling enabling it to be 0.
350 This can happen in some cases when shell redirections are used.
351 Thanks to Philippe Troin <phil@fifi.org> for the patch.
353 12. Delayed the opening of archive file until after suid had been
354 dropped (fixing a possible security issue if dump is suid).
355 Thanks to Philippe Troin <phil@fifi.org> for the patch.
357 13. Fixed the 'S' command handling in the rmt client part.
358 Thanks to Philippe Troin <phil@fifi.org> for the patch.
360 14. Modified the end-of-tape script handling to print out statistics
361 (and stop the timer) before launching the eot script. Also, the eot
362 script does not get run anymore when using -M (which makes sense) or
363 when multiple tapes are listed on the command line
364 (-f tape0,tape1,tapen) (which also makes sense).
365 Thanks to Philippe Troin <phil@fifi.org> for the patch.
367 15. Relicensed dump/restore under the 'revised' BSD license, as per
368 ftp://ftp.cs.berkeley.edu/ucb/4bsd/README.Impt.License.Change.
370 16. Added LZO compression to dump. This new compression method has
371 the advantage of being super fast, thus not killing tape streaming
372 on slow machines. Thanks to Helmut Jarausch
373 <jarausch@igpm.rwth-aachen.de> for the patch and to
374 Markus Oberhumer <markus@oberhumer.com> for giving special permission
375 to include his miniLZO project (GPL licensed) in dump/restore.
377 17. Some small buffer overruns fixes in rmt. Thanks to Antonomasia
378 <ant@notatla.demon.co.uk> for reporting the bugs.
380 18. Added a special rmt version which can do encryption when writing
381 to tape. Read examples/encrypted_rmt/README for details on
382 how to enable and configure it. Thanks to Ken Lalonde
383 <ken@globalremit.com> for the patch.
385 19. Made dump work with 2.5 kernel end of tape early warning semantics.
386 Thanks to Kai Makisara <Kai.Makisara@kolumbus.fi> for the patch.
388 20. Fixed a bug which caused dump -w|-W not to work anymore, because
389 the fs_freq and fs_passno fields in /etc/mtab are always set
390 to 0 0. Thanks to Trent Piepho <xyzzy@speakeasy.org> for
391 reporting the bug.
393 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
394 =======================================================================
396 1. Added a note in the restore man page clarifying the question
397 'set the permissions on the current directory ?' asked by
398 restore at the end of treatment in -i and -x modes.
400 2. Fixed the endianess issues when reading compressed tapes.
401 Thanks to Dark Force <daq4th@users.sourceforge.net> for
402 reporting this bug and providing test cases.
404 3. Fixed the "ACL won't be dumped" warning message (which showed
405 an extra, unrelated error message). Thanks to Dragan Krnic
406 <dkrnic@lycos.com> for reporting this bug.
408 4. Made dump look first into /etc/mtab, then into /etc/fstab
409 when searching for filesystem paths. Also fixed some problems
410 caused by binding mounts or multiple block device mounts.
411 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
412 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
413 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
414 and #160305, Sourceforge bugs #618699 and #679832).
416 5. Made dump's -I option accept the value '0' meaning all the
417 read errors will be ignored. This can be useful when running
418 dump from unattended sessions (like cron jobs). Thanks to
419 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
421 6. Fixed the output of dump to indicate 'blocks' instead of
422 'tape blocks' in the various messages (blocks are always
423 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
424 made it clearly print the current blocksize at the start of
425 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
426 the suggestions.
428 7. Made rmt understand long (largefiles) seeks.
430 8. Fixed build with very old versions of libext2fs, where
431 EXT2_FT_* constants were undefined.
433 9. Made dump accept the dumpdates path on the command line
434 (-D file option) instead of using only the hardcoded one.
435 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
436 suggestion.
438 10. Enabled rmt, LFS, readline, QFA options by default in
439 ./configure. Updated the configure process (new versions
440 of config.guess, config.sub etc).
442 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
443 =======================================================================
445 1. Changed dump to use fcntl(F_SETLK) style locking instead
446 of flock() when locking the dumpdates file. With the old
447 locking scheme, a local user having read rights on the
448 dumpdates file could be able to do a Denial of Service attack
449 on dump. In order to lock the dumpdates file with the new
450 scheme, the user would need to have write access on the file.
451 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
452 reporting the bug (originally a bugtraq post).
454 2. Fixed interactive 'ls' which caused spurious errors warnings
455 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
456 <jorgen_ostling@users.sourceforge.net> for reporting this
457 bug.
459 3. Fixed dump's estimate when dealing with sparse inodes.
461 4. Modified dump to allow setting a blocksize bigger than 32kB
462 (raised the limit to 1024kB), because newer hardware needs
463 this for proper operation. Thanks to Dirk Traenapp
464 <dtraenapp@users.sourceforge.net> for reporting this.
466 5. Fixed a bug causing Dump to stop and report an error if an
467 inode number in the exclude file was followed by some amount
468 of whitespace. Thanks to Jeffrey Sofferin
469 <sofferin@users.sourceforge.net> for reporting this bug.
471 6. Fixed a bug which caused restore, in some particular cases,
472 to ask some 'scary' questions and leave a bunch of RSTTMP
473 directories behind when restoring incremental tapes. Thanks
474 to Philippe Troin <phil@fifi.org> for reporting this bug and
475 providing the test cases.
477 7. Changed the wording when inodes are excluded from dump:
478 replaced 'Added inode 7 to exclude list' with
479 'Excluding inode 7 (resize inode) from dump', as suggested
480 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
482 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
483 ===================================================================
485 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
486 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
487 <eros@lamel.bo.cnr.it> for reporting the bug.
489 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
490 ===================================================================
492 1. Made dump print out the ext2 logical block number in case of a read
493 error, which can be given as an argument to debugfs' ncheck command,
494 in order to find out the affected inode. Added note about this
495 usage in dump's man page.
497 2. Fixed a problem in restore when reading tapes written on big
498 endian machines with very old versions of dump. The patch was
499 contributed by George Helffrich <george@geology.bristol.ac.uk>.
501 3. Fixed the tape length calculation when using large tapes
502 and compression. Thanks to Georg Lippold
503 <g_lippold@sourceforge.net> for reporting the bug.
505 4. Added a new set of examples in dump_on_cd_2 directory, based
506 on dump_on_cd examples but somewhat enhanced, supporting DVD
507 media, and localized in english and german. Thanks to
508 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
510 5. Made dump save 32 bit UID/GID. Internally, this was achieved
511 by switching from the old BSD inode format to the new BSD
512 inode format, which means that the tape format was changed.
513 However, since all restore versions out there should
514 transparently support both inode formats, the change should
515 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
516 reporting the bug.
518 6. Fixed a lot of warnings in the code shown when compiling
519 with 'gcc -W'. Thanks to Matthias Andree
520 <matthias.andree@stud.uni-dortmund.de> for reporting this.
522 7. Fixed a small markup bug in the dump man page. Thanks to
523 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
524 patch.
526 8. Rewrote entirely the man pages using the tmac.an macro
527 package (Linux man page format) instead of the original BSD
528 format. They should be now cleaner and easier to modify.
530 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
531 ==================================================================
533 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
534 non Linux clients were misinterpreted. The description of the problem
535 (incompatible numbering in Linux mtio opcodes) is documented at
536 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
537 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
538 providing an excellent, cross-platform replacement for rmt in his
539 star package.
541 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
542 which caused restore to fail when the inode maps were not entirely
543 contained on the first volume (when using really small volumes or
544 when dumping a huge number of inodes).
546 3. Fixed a problem in dump, where files in subdirectories of directories
547 that have the nodump flag set (or being excluded with -e/-E) were
548 sometimes incorrectly dumped (instead of being excluded). The original
549 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
550 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
551 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
552 help testing the patch.
554 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
555 which makes one able to pipe the output of dump, by the net, to
556 a remote CD-burner server.
558 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
559 and use GNU's symbolic syntax over rmt instead of numerical values
560 to assure multiple platform compatibility.
562 6. Documented the -d option in restore.
564 7. Added a -v (verbose) mode to dump. For now it just prints the number
565 of the inode being dumped, but this could evolve in future versions
566 to include interesting debugging output.
568 8. Added a -o flag to restore, which automatically restores the current
569 directory permissions in -i or -x mode, without asking the operator
570 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
571 and Peter Samuel <peters@e-smith.com>.
573 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
574 ====================================================================
576 1. Fixed a bug in the QFA feature which could have caused
577 a bad tape positionning on restore (causing extra delays in
578 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
580 2. Added a small note in the dump man page specifying that
581 there should be no white space between the option letter and
582 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
584 3. Made restore work with older versions of the readline library,
585 by checking in configure for several extended symbols. Restore
586 can now be compiled with a readline as old as the 2.0 release
587 (though it may be a good idea to upgrade it to a more recent
588 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
589 reporting the build failures.
591 4. Fixed a performance problem with the QFA file creation in
592 dump, which made unnecessary seeks on the tape slowing down
593 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
594 reporting this issue.
596 5. Removed the inclusion of some kernel headers in the dump
597 source, which prevented the compile in some kernel/glibc
598 headers/architecture combination. Thanks to Bdale Garbee
599 <bdale@gag.com> for reporting the bug.
601 6. Added the appropriate error message when dump fails to
602 open the output file for writing. Thanks to Amith Varghese
603 <amithv@yahoo.com> for reporting this bug.
605 7. Made restore able to understand large Solaris ufsdump tapes
606 (containing inodes bigger than 4194304). Sun have introduced
607 an "extension" to the dump tape format when dealing with
608 those inodes, which was uncorrectly handled by Linux restore.
609 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
610 providing a test case.
612 8. Added the -m parameter to dump which optimises the output for
613 inodes having been changed but not modified since the last dump
614 ('changed' and 'modified' have the meaning defined in stat(2)).
615 For those inodes, dump will save only the metadata, instead of
616 saving the entire inode contents. Inodes which are either
617 directories or have been modified since the last dump are saved
618 in a regular way. Uses of this flag must be consistent, meaning
619 that either every dump in an incremental dump set have the flag,
620 or no one has it.
622 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
623 =======================================================================
625 1. Fixed behaviour of dump when exceeding resource limits
626 (SIGXFSZ treatment).
628 2. Added the -L flag to restore to allow the user to specify a
629 maximal allowed number of miscompares when using restore
630 with the -C option to check the backup.
632 3. Detailed the manual entry for the -N option of restore.
634 4. Added the -a flag to restore to make able doing unattended
635 restores in -i or -x mode (automatically walks through the
636 multiple dump volumes).
638 5. Extended the QFA mode to work with local files and/or
639 remote tapes and files. This way, restore can know in advance
640 the tape number and the offset for the inodes to extract and
641 can minimize the extraction time by seeking directly to the
642 good tape/offset.
644 6. Added the -A <archive> option to both dump and restore,
645 which makes dump to archive a dump table-of-contents in
646 the specified file to be used by restore to determine
647 whether a file is in the dump file that is being restored.
648 (the archive file syntax is also compatible with the
649 Solaris ufsdump generated one).
651 7. Small fix in restore making it able to read some (broken ?)
652 Solaris ufsdump tapes.
654 8. Fixed dump to correctly recognise the root filesystem when using
655 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
656 for reporting this bug.
658 9. Added the -P <file> option to restore to create a
659 Quick File Access file from an already made dump. Patch
660 contributed by Uwe Gohlke <uwe@ugsoft.de>.
662 10. Made restore compile and run on Solaris, making it a
663 possible replacement for the standard ufsrestore. Port was
664 contributed by Uwe Gohlke <uwe@ugsoft.de>.
666 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
667 =====================================================================
669 1. Added a set of backup scripts from Eugenio Diaz
670 <getnito@yahoo.com> in the examples section. It features
671 automatic (cron based) full and incremental dumping of
672 several filesystems on a separate filesystem.
674 2. Fixed a off-by-one miscalculation which disabled dumping
675 a one letter subdirectory. Thanks to Chris Siebenmann
676 <cks@utcc.utoronto.ca> for reporting the bug.
678 3. Fixed several restore bugs occuring when trying to
679 restore some missing files on the tape. Thanks to Chris
680 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
682 4. Fixed --with-ldopts configure argument passing, installing from
683 a separate object directory, makefile cleanups contributed
684 by <splite@purdue.edu>.
686 5. Fix a bug which could caused, in some conditions, the highest
687 number inode of a filesystem, to not be dumped. Many thanks
688 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
689 chase this bug.
691 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
692 =======================================================================
694 1. Added a mini howto from Patrick Walsh in the examples
695 directory.
697 2. Minor man pages syntax corrections. Thanks to
698 Chris Danis <screechco@home.com> for reporting the bugs.
700 3. Added a script from David B. Peterson <dave@toppledwagon.com>
701 to the examples section. It features dumping several
702 filesystems to a remote tape drive upon ssh.
704 4. Added a patch provided by Richard Jones <rich@annexia.org>
705 which allows BRADEMAX (number of read errors tolerated by
706 dump) to be adjusted using the -I option.
708 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
709 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
710 patch.
712 6. Add the -l option to restore to specify if, when doing a
713 remote restore, the file used to access the backup is a
714 regular file (the defaults being a tape drive). Restore needs
715 to know this information when reading a remote compressed
716 dump. Previously, this information was autodetected, but
717 the autodetection code fails (with ioctl: Inappropriate ioctl
718 for device) when using a non Linux remote box. Thanks to
719 many users and especially to Eros Albertazzi
720 <eros@lamel.bo.cnr.it> for reporting this.
722 7. Found a workaroung for the dump deadlock problem (3 childs
723 stuck in pause(), father in read()). The workaround seems
724 to work for me and several beta-testers. If it doesn't work
725 for you, please report back.
727 8. Updated the RPM spec file (BuildPrereq, URL etc).
729 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
730 ========================================================================
732 1. Fixed the permissions of a newly created QFA file by dump.
734 2. Cleaned up the source of dump (the external variables
735 definition was a complete mess, making possible to have
736 objects overlap).
738 3. Fixed restore to use the full tape volume path when doing
739 a compare (since it changes the working directory to the
740 filesystem being compared in the process).
742 4. Added the -q option to dump which makes dump abort
743 whenever it needs operator attention. It should be
744 easier to use dump in scripts with this option.
746 5. Detect the use of incompatible options to dump and
747 refuse them (like -a and -B options together).
749 6. Added bzip2 compression to dump/restore (use option -j level
750 to select it). Note that you will need a restore version
751 >= 0.4b24 in order to restore a bzip2 compressed dump.
752 The same warning as for the zlib compression applies:
753 the tape format of a bzip2 dump is not compatible with the
754 original BSD tape format.
756 7. Fixed a overflow problem in dump corrupting the dump when
757 very large files were encountered. Thanks to Vladimir Nadvornik
758 for the bug report.
760 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
761 kernel buffer/page cache before starting the dump, helping
762 a bit those who use dump on mounted filesystems. Thanks to
763 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
764 for this suggestion.
766 9. Updated the RPM spec file following the RedHat changes
767 (dynamically linked binaries now in /usr/sbin etc).
769 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
770 which enables restore to recognise multi volume compressed dumps
771 done on CD. Included his scripts for dump (which pipe the dump
772 output directly into cdrecord) and restore. There is now
773 possible to to backups to CD on the fly!
775 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
776 ===================================================================
778 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
779 Marcus Meissner (Caldera International Security Dept.).
781 2. Implement the Sun rmt extended protocol. Patch submitted
782 by Ian Gordon <iangordon@users.sourceforge.net>.
784 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
785 the linux kernel header. This ensures that dump always has the
786 latest version of this file. Patch submitted by
787 Andreas Dilger <adilger@turbolinux.com>.
789 4. Report any filesystem present in either /etc/fstab with a
790 positive passno or /etc/dumpdates in dump -w output.
791 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
793 5. Fixed the looping problem in dump introduced in the
794 previous version.
796 6. Changed the -B option of dump to limit the size of
797 _compressed_ output per volume if compression is on.
798 Patch contributed by Helmut Jarausch
799 <jarausch@igpm.rwth-aachen.de>. Note however that, since
800 it is impossible to predict the size of the compressed
801 data before doing the compression, the -B limit is a bit
802 conservative.
804 7. Fixed a bug in reading the operator typed file/tape path for
805 the next volume in restore.
807 8. Implemented a "-F script" option to restore which permits the
808 user to specify a script which will be launched at the
809 beginning of each tape, useful for automatic programming of
810 tape changers for example. See the restore man page for the
811 script parameters and return codes.
813 9. Small fix for the QFA routines provided by Uwe Gohlke
814 <uwe@ugsoft.de>, and some recommendations for QFA uses in
815 the man pages.
817 10. Fixed the multivolume restoring where making a mistake
818 to the 'Mount next tape' prompt caused several blocks to
819 be lost.
821 11. Enhanced the -e option of dump to take as a parameter a
822 comma separated list of inode numbers.
824 12. Added the -E option to dump which specify a file containing
825 inode numbers to exclude from the dump.
827 13. Fixed the compressed multi-volume dump + restore.
829 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
830 ==================================================================
832 1. Made dump report the number of blocks dumped per volume.
833 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
835 2. Fix a bug in dump where waiting too much at the 'change volumes'
836 question caused the volume to be rewritten. Thanks to
837 Graham Reed <greed@users.sourceforge.net> for reporting the
838 bug and providing a patch.
840 3. Added a compression option to dump, contributed by Jerry
841 Peters <gapeters@worldnet.att.net>.
843 WARNING: the compression works only on files/pipes and
844 drives supporting variable block size.
846 WARNING2: tapes created using the compression feature are
847 incompatible with the standard dump tape format, and a
848 version of restore >= 0.4b22 will be needed for extraction.
850 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
851 architectures. Thanks to Paul Slootman <paul@debian.org> for
852 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
853 it upstream.
855 5. Many cleanups (CPP defines, const char warnings, check of
856 ext2fs COMPAT flags, time_t cleanups, added libext2 version
857 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
859 6. Made --prefix option work in configure. All the install path
860 are now based on the configure parameters.
862 7. Added the Quick File Access mode in dump/restore, contributed
863 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
864 a file tape position for each inode, and this file is used by
865 restore (if called with parameter Q and the filename)
866 to directly position the tape at the file restore is currently
867 working on. This saves hours when restoring single files from
868 large backups, saves the tapes and the drive's head. Use
869 --enable-qfa option of configure to compile in the QFA support.
871 8. Added the possibility to dump several files and directories
872 in a single invocation of dump. Thanks to Uwe Gohlke
873 <uwe@ugsoft.de> for implementing this option.
875 9. Fixed the dumping and restoring of files with holes
876 (files ending with holes didn't get dumped or restored
877 correctly).
879 10. Fixed a socket descriptor leak in dump, which leaved opened
880 3 file descriptors per dump process (and there is one dump
881 process per tape).
883 11. Fixed dump large file system support, by explicit use of
884 open64/lseek64/etc functions (explicit use needed because
885 e2fsprogs libraries don't behave well when compiled with
888 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
889 ======================================================================
891 1. Fixed some bugs in the dump -w|-W logic introduced by
892 the previous version. Thanks to Andreas Dilger
893 <adilger@turbolinux.com> for his help on this one.
895 2. Fixed again a compilation problem when using old e2fs
896 headers (filesystem label related). Thanks to many users
897 who reported this stupid error.
899 3. Fixed a build problem on old lib5 based systems dealing with
900 _PATH_MNTTAB being undefined. Thanks to John Adams
901 <johna@onevista.com> for reporting the bug.
903 4. Improved the error detection in restore when creating
904 the temporary files in TMPDIR. Now restore will corectly
905 report a 'no space left on device' error instead of
906 strange errors which could imply an archive corruption.
907 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
908 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
910 5. Added the throughput information in dump's progression
911 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
912 for the patch.
914 6. Use libext2fs's inode scanning routines, which are particularly
915 robust when dealing with errors (especially when having some
916 bad blocks on the filesystem). Many thanks to Colin
917 <colin@horizon.com> for reporting the bug, investigating
918 the issues, finding a workaround, writing the patch and
919 fully testing it... (of course, if this patch will break
920 anything, he is to blame for :-)).
922 7. Made dump and restore LFS aware. Dump can dump a filesystem
923 containing large files, generate a large file on output and
924 restore can restore them. This feature is not enabled by
925 default, use --enable-largefile option of configure to enable
926 it (you need to have a LFS aware glibc though). Thanks to
927 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
928 and to Theodore T'so <tytso@valinux.com> for his always
929 useful thoughts.
931 8. Made dump ask upon a tape write error if it should rewrite
932 the current volume (assume this is a bad tape) or if it should
933 assume an end-of-tape condition (useful for tape drives which
934 behaves badly at the end of the tape). Thanks to Andreas
935 Dilger <adilger@turbolinux.com> for the suggestion.
937 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
938 =======================================================================
940 1. Fixed a small compilation problem due to a change
941 in the definintion of the struct sigaction in
942 glibc 2.0 / libc5. Thanks to Gunther Schlegel
943 <schlegel@riege.de> for reporting the bug and to
944 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
945 a fix.
947 2. Modified the label and uuid analysis in order to be
948 self-contained (independant of kernel/libc headers). This
949 should fix the compile with older kernel/libc headers and
950 will preserve the functionality. Thanks to Bernhard Erdmann
951 <bernhard.erdmann@gmx.de> for reporting the bug.
953 3. The 'exclude inode' option, if applied to a directory
954 inode, excludes now correctly all the leaves of this
955 directory. Thanks to John R. Dennison
956 <gerdesas@users.sourceforge.net> for reporting the bug.
958 4. Fixed the '-e' option to disable the possibility
959 to exclude the root inode (which causes the dumps to
960 be unreadable by restore). Prevented array overflow
961 when multiple -e options are used.
963 5. Fixed dump to correctly interpret a filesystem argument
964 which is a mountpoint and it is not an absolute path
965 (as specified in the fstab). Thanks to Bernhard R. Erdmann
966 <be@berdmann.de> for reporting the bug.
968 6. Made dump able to backup files larger than 2 GB. Note that
969 dump still doesn't cope with files larger than 4 GB.
971 7. Restore the real uid and gid before invoking an external
972 RSH program (big hole when dump or restore is suid root!).
974 8. Get the values for the default tape device and the location
975 of fstab file from the system headers. Thanks to
976 Andreas Dilger <adilger@turbolinux.com> for the patch.
978 9. Made dump -w|-W report all recognized filesystems
979 present in either /etc/fstab or /etc/dumpdates, and present
980 the list in the same order as in fstab file. Thanks
981 to Andreas Dilger <adilger@turbolinux.com> for the patch.
983 10. Made dump's -a (automatic end of tape detection) the
984 default. Specifying one of -B, -c, -d or -s options will
985 override the EOM detection. Thanks to Andreas Dilger
986 <adilger@turbolinux.com> for the patch.
988 11. Save the ext2 filesystem label into the dump volume label.
989 Specifying a volume label on the command line (-L option)
990 will override this feature. Thanks to Andreas Dilger
991 <adilger@turbolinux.com> for the patch.
993 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
994 =====================================================================
996 1. Fixed the signal handling in dump (which I broke in 0.4b17)
997 which was causing several strange dump failures (dump
998 hanged or segmentation faults etc).
1000 2. Specified the default blocksize in the dump man page.
1002 3. Changed two info messages of restore to be written on stdout
1003 instead of stderr in order to leave stderr only for errors
1004 or warnings. Thanks to Stephen Atwell
1005 <satwell@urbana.css.mot.com> for the suggestion.
1007 4. Corrected an off by one calculation which prevented
1008 dumping a directory having a 1 character name length.
1009 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
1010 for reporting the bug.
1012 5. Reinforce dump in order to allow only level 0 and no
1013 -u option when dumping a subdirectory, and document
1014 this in the man page. Thanks to Bernhard Erdmann
1015 <bernhard.erdmann@gmx.de> for reporting the bug.
1017 6. Fixed a small harmless bug in restore which caused
1018 in some conditions hard links to be restored several
1019 times (therefore generation some warning messages).
1020 Thanks to Kenneth Porter <shiva@well.com> for
1021 reporting the bug.
1023 7. Updated the RPM spec file to the latest RedHat version,
1024 providing FHS packaging and other cosmetic changes.
1025 You will need RPM version >= 3.0.5 in order to build the RPM.
1027 8. Updated the configure script to check for libtermcap
1028 before checking for libreadline, because we need this
1029 library in order to compile the readline support.
1031 9. Made dump understand the LABEL= and UUID= notation
1032 both in /etc/fstab and on the command line. Note that
1033 you will need the /proc filesystem in order to use
1034 these notations. Thanks to Erik Troan <ewt@redhat.com>
1035 for providing the patch.
1037 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
1038 ===================================================================
1040 1. Fixed a potential buffer overflow in restore. Thanks
1041 to Stan Bubrouski <satan@fastdial.net> for reporting
1042 the bug.
1044 2. Fixed a readline-related bug which prevented
1045 'cat DUMPFILE | restore -i -f -' from working. Thanks
1046 to Charles Karney <karney@users.sourceforge.net>
1047 for the bug report.
1049 3. Changed a few "panic" into "exit", causing restore to
1050 be more stable against some attacks (like the last one
1051 reported on Bugtraq, although the last version of restore
1052 was not vulnerable - just dumped core). Thanks to
1053 Andreas Hasenack <andreas@conectiva.com.br> for reporting
1054 the bugs to me.
1056 4. Removed the suid-root bit on dump and restore in the
1057 default build (and generated RPMs). It should be safer
1058 now. Users who need the suid-root capabilities in order
1059 to make network backups should read first the man page
1060 of dump and enable the suid bit manually.
1062 5. Added -ltermcap to the compile parameters for restore
1063 when using readline support, in order to make the compile
1064 process work on some systems (SuSE ?). Thanks to
1065 Patrik Schindler <poc@pocnet.net> for reporting the bug.
1067 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
1068 ====================================================================
1070 1. The -F script is called now *only* at the end of a tape,
1071 not at the end of the dump. Thanks to Christian Haul
1072 <haul@informatik.tu-darmstadt.de> for the bug report.
1074 Normally, the device name and the current volume number
1075 are passed on the command line of the script. However,
1076 if you want the old style script (with no arguments
1077 passed on the command line), you can enable it in
1078 configure with the --enable-oldstylefscript.
1080 2. Use posix signal handling to preserve dump functionality
1081 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
1082 the patch.
1084 3. Made the exit code of restore in case of a 'C'ompare
1085 command reflect the result of the compare. An exit status
1086 of 0 means the dump archive is correct, 1 means tape errors,
1087 2 means that some files were modified. Thanks to Kenneth Porter
1088 <shiva@well.com> for the suggestion.
1090 4. Made (finally) quotes work in interactive restore.
1092 5. Small fixes in order to allow dump to compile with a
1093 really old e2fsprogs version. Thanks to Ian Zimmerman
1094 <itz@speakeasy.org> for the bug report.
1096 6. Add GNU readline capabilities to interactive restore.
1097 Use configure's --enable-readline flag to enable this feature.
1098 Thanks to Patrik Schindler <poc@pocnet.net> for the
1099 suggestion.
1101 7. Do the compare on the fly in restore 'C' mode (this will
1102 allow not to exhaust the available /tmp space when
1103 ccmparing large files). Thanks to Kenneth Porter
1104 <shiva@well.com> for the suggestion.
1106 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
1107 ====================================================================
1109 1. Fixed some several duplicate 'DUMP: DUMP:' in the
1110 output of dump.
1112 2. Corrected the estimation of blocks to dump. Note that
1113 this is still wrong for large directory inodes, due
1114 to the size of a BSD directory entry on the tape
1115 (estimation is based on the size of the ext2 directory,
1116 which is smaller because it doesn't contain the null
1117 character at the end).
1119 3. Made dump report the total number of MB written to
1120 tape at the end of run. Thanks to W. Reilly Cooley
1121 <wcooley@nakedape.cc> for the patch.
1123 4. Added the -X option to restore giving the possibility
1124 to read the names of the files to be extracted/listed
1125 from a text file (in addition of the command line).
1126 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
1127 patch.
1129 5. Added the device name and the current volume number
1130 as arguments to the end of tape script (dump -F option).
1132 6. Made the multi-volume dump work again (after having
1133 broken it in 0.4b15).
1135 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
1136 ===================================================================
1138 1. Added a prompt command in interactive restore mode. Thanks
1139 to Andreas Dilger <adilger@home.com> for the patch.
1141 2. Fixed a buffer overflow problem in dump (caused by
1142 not checking the size of the filesystem parameter).
1143 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
1144 reporting this on Bugtraq (and to several dump users
1145 who forwarded me his mail).
1147 3. Added the '-F script' option to dump in order to
1148 launch a script at the end of each tape (to be used
1149 with a tape changer, or to notify the sysadmin by
1150 pager etc.).
1152 4. Fixed a bug in restore compare code caused by the changes
1153 I made in 0.4b14.
1155 5. Fixed the treatment of options using the old BSD syntax
1156 in both dump and restore.
1158 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
1159 =======================================================================
1161 1. Fixed a bug in dump which may have caused invalid deleted
1162 directories to be dumped out if they were deleted after the
1163 mapping pass. This could occure on active filesystem and lead
1164 to heap corruption (causing dump malfunction in many possible ways).
1165 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
1166 bug and submitting the patch.
1168 2. Fixed the handling of the filesystem parameter in dump. A
1169 '/mnt/disk' parameter caused the disk contents to be dumped,
1170 but a '/mnt/disk/' parameter caused the mountpoint directory
1171 to be dumped (generally an empty directory).
1173 3. Improved the output of dump in order to tell which directory
1174 it is currently dumping (when dumping a subtree).
1176 4. Added the '-e' exclude inode option to dump. Thanks to
1177 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
1179 5. Added a REPORTING-BUGS file in order to provide a guide
1180 on how to correctly report a bug in dump/restore.
1182 6. Fixed a restore problem when restoring a hard link to an inode
1183 having the immutable or append-only attribute set. Thanks to
1184 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
1186 7. Fixed a compatibility problem between dumps made on little
1187 endian machines (the format was correct) and big endian
1188 machines (the format was incorrect). This fix break the
1189 compatibility with the older dumps made on big endian
1190 machines (sparc, m86k, ppc etc). For the first time in
1191 linux dump's history (I believe), the dumps made by *BSD,
1192 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
1193 so interchangeable. Thanks to Rob Cermak
1194 <cermak@ahab.rutgers.edu> for submitting the bug and
1195 helping me test the fix.
1197 8. Fixed the way dump reports the remaining percent/time, if
1198 the number of blocks actually dumped exceeds the estimated
1199 number of blocks. Thanks to Jean-Paul van der Jagt
1200 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
1202 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
1203 ======================================================================
1205 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
1206 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
1208 2. Removed the requirement to build the RPM as root from the
1209 spec file. Thanks to Christian Weisgerber
1210 <naddy@mips.rhein-neckar.de> for submitting this.
1212 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
1213 causing dump to block when dumping really huge filesystems.
1214 Many thanks to Patrik Schindler <poc@pocnet.net> for
1215 helping me find this bug.
1217 4. Fixed the treatment for an interrupt signal when dump access
1218 the remote tape through RSH. Thanks to Christian Weisgerber
1219 <naddy@mips.rhein-neckar.de> for providing the patch.
1221 5. Fixed a bug which was causing dump/restore to display
1222 garbage characters instead of the remote host name.
1224 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
1225 =====================================================================
1227 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
1228 <kukuk@suse.de> for submitting the patch.
1230 2. Fix for the exit code when using the size estimate option of
1231 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
1232 submitting the patch.
1234 3. Handle EINTR in atomical reads/writes in dump, which was causing
1235 dump to fail on some systems. Thanks to Eric Jergensen
1236 <eric@dvns.com> for reporting the bug and submitting the patch.
1238 4. Handle more than 16 characters for the device names in dumpdates.
1239 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
1240 tracking down the problem and proposing the solution.
1242 5. Fixed a bug in dump which prevented the creation of the
1243 dumpdates file when doing a 0-level dump without already
1244 having a dumpdates file. Thanks to Patrik Schindler
1245 <poc@pocnet.net> for reporting the bug.
1247 6. Changed the way dump 'S' flag reports the size estimate
1248 from number of blocks into bytes (making it compatible
1249 with the Solaris version, and simplifying things for
1250 amanda users). Thanks to Jason L Tibbitts III
1251 <tibbs@math.uh.edu> for reporting the bug.
1253 7. Fixed a compatibility problem in linux/alpha dump tape format.
1254 Now the linux/alpha dump are (again) compatible with the
1255 other dump formats. But this breaks compatibility with
1256 older dumps made on alpha. Thanks to Mike Tibor
1257 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
1259 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
1260 ======================================================================
1262 1. Added a '--enable-kerberos' to configure.
1264 2. Added a 'S' option to dump which determines the amount of space
1265 that is needed to perform the dump without actually doing it, similar
1266 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
1267 <cermak@ahab.rutgers.edu>.
1269 3. Added a 'M' multi-volume option to dump and restore which enables
1270 dumping to multiple files (useful when dumping to an ext2
1271 partition to make several dump files in order to bypass the 2GB
1272 file size limitation). The 'f' argument is treated as a prefix and
1273 the output files will be named <prefix>001, <prefix>002 etc. With
1274 the 'M' flag, restore automatically selects the right file without
1275 asking to enter a new tape each time.
1277 4. Fixed a memory leak which was causing dump to grow very big
1278 (270MB when dumping a 10GB filesystem...). Thanks to Jason
1279 Fearon <jasonf@netrider.org.au> for reporting the bug.
1281 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
1282 ======================================================================
1284 1. Make configure test if the system glob routines support
1285 extended syntax (ALTDIRFUNC). If not, use the internal glob
1286 routines instead of system ones. Thanks to Bernhard Sadlowski
1287 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
1288 and helping me resolve this and other minor libc5 compiling
1289 glitches.
1291 2. Fix a problem when dumping a ext2fs with the 'filetype'
1292 feature enabled. Thanks to Patrick J. LoPresti
1293 <patl@cag.lcs.mit.edu> for reporting the bug and to
1294 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
1296 3. Made the nodump flag work on directories. A directory which
1297 has the nodump flag gets never dumped, regardless of its
1298 contents.
1300 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
1301 which allows dump on an active ext3 filesystem. However, this
1302 is a "quick and dirty" patch which enables backup of an ext3
1303 filesystem through the ext2 compatibility (by ignoring the
1304 NEEDS_RECOVERY bit). The journal file is not recognized and
1305 it is dumped (it should not).
1307 5. Test the superblock compatibility flags when dumping, in order
1308 to be sure that we know how to deal with specific features.
1310 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
1311 ====================================================================
1313 1. Use lchown instead of chown, fixing a possible security problem
1314 when restoring symlinks (a malicious user could use this
1315 to deliberately corrupt the ownership of important system files).
1316 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
1317 this and providing the patch.
1319 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
1320 ====================================================================
1322 1. Put dump sources under CVS, added Id tags in all files so
1323 one can use 'ident' on binary files.
1325 2. Added the dump/restore version in the usage text so one can
1326 easily verify the version he is using.
1328 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
1329 a va_start/va_end problem on linux-ppc (always call va_start
1330 va_end in pairs each time we use a vararg function).
1332 4. Added again the DT_* constants because old libc does not
1333 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
1334 for submitting the bug report.
1336 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
1337 this should enable dumping big (huge) filesystems.
1339 6. Added the RSH environment variable in order to be able to
1340 use a rsh replacement like ssh when doing remote backups (and
1341 bypass the security limitations of rcmd). Now you can do remote
1342 backups without being root (or making dump setuid root).
1344 7. Modified again the way dumpdates works. For incremental dumps,
1345 we need to read dumpdates even if we are not using 'u' option.
1346 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
1347 this should work.
1349 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
1350 ===================================================================
1352 1. Removed the 'k' flag from the restore 'about' text if kerberos
1353 was not compiled in.
1355 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
1356 to them (fsetflags takes a char*, setflags an open fd!).
1358 3. (f)setflags is called only if the flags aren't empty. If the
1359 file is a special file, a warning is printed, because changing
1360 flags implies opening the device. Normally, a special file
1361 should not have any flag... (Debian bug #29775, patch provided
1362 by Abhijit Dasgupta <abhijit@ans.net>).
1364 4. Made possible to dump a file system not mentioned in /etc/fstab.
1365 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
1367 5. Changed the default behaviour to not create dumpdates
1368 unless 'u' option is specified. Removed the old "debian-patch"
1369 which provided the same thing. (Debian bug #38136, #33818).
1371 6. Removed all those dump*announce, since they were getting old...
1373 7. Added warning messages if dumpdates does not exist and
1374 when an inferior level dump does not exist (except for a level 0
1375 dump).
1377 8. Debugged the glob calls in interactive mode: restore used a
1378 dirent struct which was different from the /usr/include/dirent.h
1379 one (this used to work, is it a glibc2 change?), so none of the
1380 compat glob (which used /usr/include/dirent.h) or the system glob
1381 worked. Restore use now the system dirent (and the system
1382 DT_* constants), which are compatible with BSD ones.
1384 9. Added a configure flag (--with-dumpdatespath) to specify
1385 the location of dumpdates file. By default, it is
1386 /etc/dumpdates.
1388 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1389 included the current date/version in man pages.
1391 11. Corrected the estimation of remaining time when
1392 the operator doesn't change the tapes quickly enough. This
1393 was an old bug, I thought I corrected it, and discovered
1394 that in fact it was corrected in two different places, so
1395 the results canceled each other...
1397 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1398 ===================================================================
1400 1. Integrated multiple patches from RedHat, Debian and SuSE:
1402 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1403 isn't specified.
1404 - several fixes in the man pages.
1405 - update the default tape device to /dev/st0.
1406 - many updates for Linux Alpha (byte ordering, size_t etc).
1407 - buffer overruns.
1408 - use environment variable for TMPDIR (instead of /tmp).
1409 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1410 - workaround egcs bug (RedHat bugs #4281 and #2989).
1411 - wire $(OPT) throughout Makefile's.
1413 2. Upgrade the dump revision to 1, making possible to dump filesystems
1414 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1416 3. Fix some compile warnings, prototype all functions.
1418 4. Use glibc err/glob instead of internal compatibility
1419 routines (only if available).
1421 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1422 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1424 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1425 ======================================================================
1427 1. Integrated the changes from FreeBSD-3.1-RELEASE
1428 (mostly bug fixes, buffer overruns, dump has now an "automatic
1429 tape length calculation" flag, dump/restore can use kerberos now
1430 (this is NOT tested), use environment variables for TAPE and
1431 RMT etc.).
1433 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1435 3. Corrected a bug in restore when using 'C' option with multi-volumes
1436 tapes (files splited accros two tapes give "size changed" errors
1437 when comparing).
1439 4. Corrected the long standing bug when dumping multiple tapes.
1440 This works for me, needs further testing.
1442 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1443 ====================================================================
1445 1. Dump now runs correctly on kernels 2.1.x
1446 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1448 Changes between versions 0.4b2 and 0.4b3
1449 ========================================
1451 1. Use realpath() if available
1453 2. Report statistics
1455 Changes between versions 0.4b1 and 0.4b2
1456 ========================================
1458 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1459 the patch)
1461 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1463 3. Fixed configure.in to correctly handle --enable-rmt
1465 Changes between versions 0.3 and 0.4b1
1466 ======================================
1468 1. Integrated the changes from 4.4BSD-Lite2
1470 2. Integrated the patches from Debian and Red Hat
1472 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1474 4. Changed dump to use the Ext2fs library to get block addresses. This
1475 should solve the endianness problem on SparcLinux.
1477 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1478 distribution's one) to use autoconf
1480 6. Fixed a few minor bugs
1482 Changes between versions 0.2e and 0.2f
1483 ======================================
1485 1. Added the creation of named pipes (fifos) in restore.
1487 2. Added the -N flag in the restore manual page.
1489 3. Added the file kernel.patch which contains the llseek() optimization
1490 patch for 1.2.x kernels.
1492 4. Fixed a bug in the restoration of symbolic links: owner and group were
1493 not restored.
1495 5. Integrated some changes from FreeBSD 2.2.
1497 6. Added a call to ftruncate() after restoring each file to restore
1498 correctly files ending by a hole.
1500 Changes between versions 0.2d and 0.2e
1501 ======================================
1503 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1504 with owner = root (0) and group = root/wheel/whatever (0).
1506 Changes between versions 0.2c and 0.2d
1507 ======================================
1509 1. Dump is now able to backup 2GB+ filesystems.
1511 2. Dump and restore can now be linked as static binaries.
1513 Changes between versions 0.2b and 0.2c
1514 ======================================
1516 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1518 Changes between versions 0.2a and 0.2b
1519 ======================================
1521 1. Really fixed the bug that I should have corrected in 0.2a.
1523 2. Enabled optimization again.
1525 Changes between versions 0.2 and 0.2a
1526 =====================================
1528 1. Disabled the optimization during compilation.
1530 Changes between versions 0.1 and 0.2
1531 ====================================
1533 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1534 the fs_type field (actually, I modified the file fstab.c to make it
1535 use the mntent functions).
1537 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1538 made by dump should be readable by the BSD restore and backups made
1539 by the BSD dump should be readable by restore. Unfortunately, this
1540 means that the dump format has changed between version 0.1 and version
1541 0.2 :-(
1543 3. Dump is now able to backup a subtree, it is no longer limited to whole
1544 filesystems like the BSD version.
1546 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1547 than 2 GB.
1549 Changes between versions 0.0 and 0.1
1550 ====================================
1552 1. Now create links rdump and rrestore during the `make install' step.
1554 2. Linux port specific bugs added to the manual pages
1556 3. Incorrect estimation of the number of tapes blocks fixed when doing
1557 incremental backups.
1559 4. Better ls-like format in restore in interactive mode.