1 $Id: CHANGES,v 1.303 2009/07/23 09:34:07 stelian Exp $
3 Changes between versions 0.4b42 and 0.4b43 (released ?????????????)
4 ===================================================================
6 1. Fix a bug in dump making impossible to handle large toc files
7 (> 2 GB). Thanks to X DUGi <xdugi@users.sourceforge.net> for
8 reporting the bug (Sourceforge bug #2820629)
10 Changes between versions 0.4b41 and 0.4b42 (released June 18, 2009)
11 ===================================================================
13 1. Fix printout of the 'Connection to' message in dump/restore.
14 Thanks to Dale Wiles <dwiles@users.sourceforge.net> for the
15 bug report and its fix.
17 2. Fix dump -w/-W output for filesystems having the last backup
18 level equal to 0. Thanks to Pascal Bouchareine
19 <kalou@users.sourceforge.net> for reporting the bug.
21 3. Enable restore to handle restoring onto a different SELinux
22 policy from the dump. Thanks to Tony Nelson
23 <tony_nelson@users.sourceforge.net> for the patch.
24 (see http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=189845
27 4. Fix EA set failures when restoring immutable files. Thanks
28 to Andrew Kroeger for the patch.
30 5. Fix "mode file too large" errors when restoring huge backups.
31 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
34 6. Add the ability to link against libtinfo library if present,
35 instead of libtermcap or libncurses libraries. Thanks to
36 Adam Tkac <atkac@redhat.com> for the patch.
38 7. When comparing a backup, do not consider to be an error the fact
39 that we encounter files created while dump was in progress.
40 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
43 8. Force '-a' to be the default value when running 'restore -C'.
44 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
47 9. Many improvements to the 'cron_dump_to_disk' example.
48 Thanks to Aaron S. Hawley <ashawley@users.sourceforge.net> for
51 10. Fix the QFA generation when extended attributes are backuped
52 along with the inodes. Prior to this fix, the entries for
53 some inodes may get corrupted (duplicated entries or incorrect
54 ones), making the QFA file unusable for repositionning in
55 restore. Many thanks to Kenneth Porter <shiva@sewingwitch.com>
56 for the bug report and the testing of the fix.
58 11. Fix EA restoration in debug mode (Sourceforge bug #1986216).
60 12. Stefan Auracher <star7@onlinehome.de> noticed that the Tower
61 of Hanoi backup strategy documented in the man page wasn't
62 actually based on the Tower of Hanoi algorithm. The man page
63 was updated accordingly.
65 13. Fix missing level information from dump output when using
66 the default level (RedHat bug #493635). Thanks to Adam Tkac
67 <atkac@redhat.com> for the patch.
69 14. Fix the dump man page to reflect the fact that the default
70 dump level is 0 not 9 (RedHat bug #356121). Thanks to Adam Tkac
71 <atkac@redhat.com> for the patch.
73 15. Fixed a few spelling errors in the man pages (RedHat bug #489853).
74 Thanks to Adam Tkac <atkac@redhat.com> for the patch.
76 16. Fixed restore man page and program usage which incorrectly stated
77 that -P and -A were both allowed in the same invocation (RedHat
78 bug #490627). Thanks to Adam Tkac <atkac@redhat.com> for the patch.
80 17. Use sys_clone under Linux to share I/O contexts between dump
81 processes, thus drastically increasing the performance of dump
82 under CFQ (which is the default I/O scheduler used in a number of
83 distributions). Thanks to Jeff Moyer <jmoyer@redhat.com> for the
86 18. Add (preliminary) ext4 support - thanks to libext2fs which does
87 all the job for us. Thanks to Gertjan van Wingerde
88 <gwingerde@gmail.com> for the patch.
90 Changes between versions 0.4b40 and 0.4b41 (released January 2, 2006)
91 =====================================================================
93 1. Fix restore of dumped Access Control Lists. The previous
94 code for EA works fine, but ACLs needed conversion from the
95 ext2/3 disk format to posix_acl format before restoring.
97 2. Fix some issues with restoration of EA on big endian
100 3. Fix restore when the symtab is over 2GB in size.
102 4. Made the directory hash indexing an optional feature,
103 accessible by the '-H' option of restore, and disable it
106 5. Fixed dump to not include extended attributes information
107 in the toc (archive) file which confused restore -t.
109 Changes between versions 0.4b39 and 0.4b40 (released May 2, 2005)
110 =================================================================
112 1. Changed restore to emit warnings (instead of emitting a fatal
113 error) if a file (or a directory) is unavailable for a
114 comparision (if the user doesn't have the necessary permissions
115 to access it for example). Thanks to Kenneth Porter
116 <shiva@sewingwitch.com> for the bug report.
118 2. Re-done the 'do not save directory entries to non-dumped inodes
119 (excluded from dump)' feature. The previous implementation
120 worked well for excluded directories but not for regular files.
121 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
124 3. Fixed a bug in dump where the tape size was miscalculated when
125 the user used -d/-s to specify the tape characteristics. Thanks
126 to Philip Goisman <goisman@physics.arizona.edu> for reporting
129 4. Fixed another bug introduced in restore with the hashtree
130 implementation. This one caused restore to stop saying
131 "removenode: non-empty directory" in some cases.
133 5. Added support for dumping and restoring ext2/3 extended
134 attributes (EA), like the access control lists (ACL) or
135 the security labels used by SELinux.
137 Changes between versions 0.4b38 and 0.4b39 (released January 21, 2005)
138 ======================================================================
140 1. The newly added dump_on_cd_3 example was buggy, replace it
141 with an updated version from Andrew Basterfield
142 <bob@cemetery.homeunix.org>.
144 2. Made restore to chdir() back into the initial directory when
145 dumping core while aborting a comparision operation. The
146 previous behaviour was to write the corefile at the root of
147 the directory being compared, which could very well be
148 read only and preventing the corefile generation. Thanks
149 to Kenneth Porter <shiva@sewingwitch.com> for the bug report.
151 3. Silenced the failure to call fgetflags() when comparing an
152 entry which has no ext2 attributes (as in lsattr()).
154 4. Fix a brown paper bug in restore -C which broke restore and
155 caused modifications on the filesystem being compared
156 (directories containing a file with the same name as the
157 directory get renamed to RSTTMP...). Thanks to Kenneth Porter
158 <shiva@sewingwitch.com> for finding the bug and helping me
161 5. Made restore -C force the -N flag (no writing allowed on
162 the disk) in order to prevent more bugs like the above one.
164 Changes between versions 0.4b37 and 0.4b38 (released January 7, 2005)
165 =====================================================================
167 1. Fix a couple of troff syntax bugs in the man pages.
168 Thanks to Eric Raymond <esr@thyrsus.com> for the patch.
170 2. Made restore use either libncurses or libtermcap, depending
171 on which one is available at configure time.
173 3. Fixed restore negative size display bug when comparing a
174 dump containing files over 2GB. Thanks to Steve Bonds
175 <sbonds@users.sourceforge.net> for the bug report.
177 4. Do not save directory entries to non-dumped inodes
178 (excluded from dump). This will eliminate the 'missing
179 file' warnings when doing 'restore -C'.
181 5. Fix dump crash when backuping a huge (2TB) filesystem,
182 due to a bogus calculation on the inode map size.
183 Thanks to Kevin B. Haines <K.B.Haines@rl.ac.uk> for
184 submitting the bug and testing the fix.
186 6. Fix a problem in restore where the final \0 in the symbolic
187 link names could have been lost, generating corrupt filenames.
188 Thanks to Kyle Wilson <kyle.wilson@amd.com> for reporting the
191 7. Implemented a hash list for the directory names in restore.
192 The linear list used before caused problems in interactive
193 restores when dealing with directories having thousands of
194 entries. Thanks to Brian Ristuccia <bristuccia@starentnetworks.com>
195 for reporting the bug.
197 8. Improved restore -C, this time including the directory
198 attributes into the comparision.
200 9. Made restore understand tapes containing EA/ACLs (which will
201 be dumped by the next version of dump). In this version
202 extended attributes on the tape are ignored, for full EA/ACL
203 support wait for the next version or try the experimental EA
206 Changes between versions 0.4b36 and 0.4b37 (released July 7, 2004)
207 ==================================================================
209 1. Added the --enable-staticz configure option which enables
210 dump and restore to be linked with static versions of
211 libz and libbz2 (and dynamic versions of all other libraries).
212 This will make Debian users happy, because libz and libbz2
213 were the only needed libraries living in /usr, all the
214 others live in /lib. In case of system emergency, it is
215 better not to have to rely on an extra filesystem.
216 Thanks to Bdale Garbee <bdale@gag.com> for the suggestion.
218 2. Fix compilation on (at least the Linux Debian port to) AMD64.
219 (<ext2fs/ext2_types.h> defines some types (__s64 and __u64)
220 that are also defined by <linux/types.h> (<asm/types.h>) and
223 3. Make dump's reading of the dumpdates file a bit more robust,
224 preventing dump from crashing when the dumpdates file has
225 been modified by hand.
227 4. Fixed some offset calculations in dump code which could
228 lead to "bread lseek errors" on large filesystems. Thanks
229 to Bruce Lowekamp <lowekamp@users.sourceforge.net> for
230 reporting this bug and debugging the issue.
232 5. Made dump use the blkid library when searching for devices
233 by label or uuid instead of dump's own routines.
235 6. Corrected a bug in dump where a wrong LABEL=... line in
236 /etc/fstab could prevent dump from dumping unrelated
237 filesystems. Thanks to Bruce Lowekamp
238 <lowekamp@users.sourceforge.net> for reporting the bug.
240 Changes between versions 0.4b35 and 0.4b36 (released April 21, 2004)
241 ====================================================================
243 1. Fixed dump compilation with old gcc versions. Thanks to
244 Mike Castle <dalgoda@users.sourceforge.net> for the patch.
246 2. Fixed some warnings (howmany, roundup, powerof2 redefined)
247 when compiling against a recent glibc version.
249 3. Fixed a bug in restore preventing the read of a dump tape
250 written with Solaris 7 ufsdump. Thanks to Patrick Higgins
251 <phiggins@transzap.com> for reporting the bug and providing
254 4. Changed dump to enable the creation of volumes of different
255 sizes in a single run (make -B accept a list of values).
256 Patch contributed by Florian Zumbiehl <florz@gmx.de>.
258 5. Use the glibc provided minor() and major() macros instead
259 of our own bitmask implementation. This should be safe for
260 when the major/minor namespace will migrate to 32 bits.
261 Thanks to Zhang Jun <zhangjun@nanjing-fnst.com> for reporting
264 6. Made explicit in the dump man page that dump will not create
265 a remote file, it will only write to an already existing one.
267 7. Another try at making size estimates better again.
269 8. Put back the inconditional running of the end-of-tape
270 script which was changed in 0.434 to be run only when -M
271 or multiple -f were NOT specified. Some users rely on this
272 feature even when it is combined with -M/-f.
274 9. Fixed restore when restoring huge backups (where rstdir...
275 temporary files are over 2GB). Thanks to
276 Raphael Jaffey <rjaffey@artic.edu> for reporting this,
277 debugging the issue and making the patch.
279 10. Made restore understand FreeBSD UFS2 tapes. Thanks to
280 David <vrtin@users.sourceforge.net> for submitting the bug
281 and providing a test case.
283 11. Made dump work with an arbitrary integer level (no more 10
284 levels only limitation). Thanks to Matthew
285 <msvincen@midway.uchicago.edu> for the patch.
287 Changes between versions 0.4b34 and 0.4b35 (released December 21, 2003)
288 =======================================================================
290 1. Added a note in the dump man page saying that the default
291 blocksize can be 32 if -d is used with a high density argument.
292 Thanks to Antonios Christofides <A.Christofides@itia.ntua.gr>
295 2. Fixed configure to correctly understand CPPFLAGS, CFLAGS,
296 LDFLAGS environment variables. Thanks to Arcady Genkin
297 <antipode@users.sourceforge.net> for reporting the bug.
299 3. Made -e/-E options of dump accept an unlimited number of inodes
300 to be excluded, rather than a hardcoded maximum. Thanks to
301 Dietrich Rothe <d-rothe@users.sourceforge.net> for the patch.
303 4. Updated the autoconf system to 2.50. Forced the -D_BSD_SOURCE
304 and -D_USE_BSD_SIGNAL defines in configure in order to solve
305 64bit build problems because quad_t is redifined with a
306 different signature. Thanks to Mike Harris <mharris@redhat.com>
307 for reporting this bug.
309 5. Made restore build on Solaris, making possible to
310 restore Linux's "enhanced" tapes. Thanks to Uwe Gohlke
311 <uwe@ugsoft.de> for the patch.
313 6. Made an extension in the dump tape format capable of saving
314 MacOSX specific inode extensions. Uwe Gohlke <uwe@ugsoft.de>
315 wrote the extension and contributed the restore code back
316 into this codebase. The same extension mechanism will be
317 used in the future to save ACLs...
319 7. Made rmt work correctly with regard to QFA and local/remote
320 files and tapes. The remote access will however work only
321 when the dump provided rmt version is used. If you want to
322 use another rmt server, please do not use the QFA feature.
323 Thanks to Uwe Gohlke <uwe@ugsoft.de> for the patch.
325 Changes between versions 0.4b33 and 0.4b34 (released April 18, 2003)
326 ====================================================================
328 1. Fixed the INSTALL file to reflect the actual install paths.
329 Thanks to David Ronis <ronis@ronispc.chem.mcgill.ca> for
332 2. Fixed the configure script to only check for headers presence
333 instead of trying to compile them. This should fix issues
334 with old build environments. Thanks to Kari Lempiainen
335 <kari@funky.pp.fi> for reporting the bug.
337 3. Fixed restore to correctly ignore sockets when comparing
338 a dump (as socket cannot be properly restored anyway). Thanks
339 to Gunther Reiszig <gunther@mit.edu> for reporting the bug.
341 4. Fixed restore to correctly access the archive file (-A argument)
342 even when using a remote tape. Thanks to Fabrice Bellet
343 <fabrice@bellet.info> for reporting the bug.
345 5. Fixed (again) handling of long (largefile) seeks in rmt.
346 Thanks to Fabrice Bellet <fabrice@bellet.info> for reporting
349 6. Fixed restore corner case when dealing with large block sizes
350 dump is able to write now (-b 1024). Thanks to Fabrice Bellet
351 <fabrice@bellet.info> for reporting the bug.
353 7. Fixed a bug preventing dump to access a filesystem having
354 a label exactly 16 bytes in length. Thanks to <minduni@ti-edu.ch>
355 for reporting the bug.
357 8. Made dump store dump dates using explicit timezones, fixing a
358 problem with incremental dumps if the timezone is modified
359 between the dumps. Thanks to Philippe Troin <phil@fifi.org> for
360 the bug report and the patch.
362 9. Fixed a bug encountered when dumping individual files (not full
363 filesystems or directories) and dangling symbolic links happen
364 to be in the list of files. For as far as dump is concerned,
365 dangling symbolic links are allowed, and are dumped as is.
366 Thanks to Jin-su Ahn <jsahn@ee.snu.ac.kr> for reporting the
367 bug and providing the fix.
369 10. Fixed open and creation modes and permissions for QFA and
370 table-of-contents files in dump and restore. Thanks to
371 Philippe Troin <phil@fifi.org> for the patch.
373 11. Fixed the archive file descriptor handling enabling it to be 0.
374 This can happen in some cases when shell redirections are used.
375 Thanks to Philippe Troin <phil@fifi.org> for the patch.
377 12. Delayed the opening of archive file until after suid had been
378 dropped (fixing a possible security issue if dump is suid).
379 Thanks to Philippe Troin <phil@fifi.org> for the patch.
381 13. Fixed the 'S' command handling in the rmt client part.
382 Thanks to Philippe Troin <phil@fifi.org> for the patch.
384 14. Modified the end-of-tape script handling to print out statistics
385 (and stop the timer) before launching the eot script. Also, the eot
386 script does not get run anymore when using -M (which makes sense) or
387 when multiple tapes are listed on the command line
388 (-f tape0,tape1,tapen) (which also makes sense).
389 Thanks to Philippe Troin <phil@fifi.org> for the patch.
391 15. Relicensed dump/restore under the 'revised' BSD license, as per
392 ftp://ftp.cs.berkeley.edu/ucb/4bsd/README.Impt.License.Change.
394 16. Added LZO compression to dump. This new compression method has
395 the advantage of being super fast, thus not killing tape streaming
396 on slow machines. Thanks to Helmut Jarausch
397 <jarausch@igpm.rwth-aachen.de> for the patch and to
398 Markus Oberhumer <markus@oberhumer.com> for giving special permission
399 to include his miniLZO project (GPL licensed) in dump/restore.
401 17. Some small buffer overruns fixes in rmt. Thanks to Antonomasia
402 <ant@notatla.demon.co.uk> for reporting the bugs.
404 18. Added a special rmt version which can do encryption when writing
405 to tape. Read examples/encrypted_rmt/README for details on
406 how to enable and configure it. Thanks to Ken Lalonde
407 <ken@globalremit.com> for the patch.
409 19. Made dump work with 2.5 kernel end of tape early warning semantics.
410 Thanks to Kai Makisara <Kai.Makisara@kolumbus.fi> for the patch.
412 20. Fixed a bug which caused dump -w|-W not to work anymore, because
413 the fs_freq and fs_passno fields in /etc/mtab are always set
414 to 0 0. Thanks to Trent Piepho <xyzzy@speakeasy.org> for
417 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
418 =======================================================================
420 1. Added a note in the restore man page clarifying the question
421 'set the permissions on the current directory ?' asked by
422 restore at the end of treatment in -i and -x modes.
424 2. Fixed the endianess issues when reading compressed tapes.
425 Thanks to Dark Force <daq4th@users.sourceforge.net> for
426 reporting this bug and providing test cases.
428 3. Fixed the "ACL won't be dumped" warning message (which showed
429 an extra, unrelated error message). Thanks to Dragan Krnic
430 <dkrnic@lycos.com> for reporting this bug.
432 4. Made dump look first into /etc/mtab, then into /etc/fstab
433 when searching for filesystem paths. Also fixed some problems
434 caused by binding mounts or multiple block device mounts.
435 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
436 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
437 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
438 and #160305, Sourceforge bugs #618699 and #679832).
440 5. Made dump's -I option accept the value '0' meaning all the
441 read errors will be ignored. This can be useful when running
442 dump from unattended sessions (like cron jobs). Thanks to
443 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
445 6. Fixed the output of dump to indicate 'blocks' instead of
446 'tape blocks' in the various messages (blocks are always
447 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
448 made it clearly print the current blocksize at the start of
449 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
452 7. Made rmt understand long (largefiles) seeks.
454 8. Fixed build with very old versions of libext2fs, where
455 EXT2_FT_* constants were undefined.
457 9. Made dump accept the dumpdates path on the command line
458 (-D file option) instead of using only the hardcoded one.
459 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
462 10. Enabled rmt, LFS, readline, QFA options by default in
463 ./configure. Updated the configure process (new versions
464 of config.guess, config.sub etc).
466 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
467 =======================================================================
469 1. Changed dump to use fcntl(F_SETLK) style locking instead
470 of flock() when locking the dumpdates file. With the old
471 locking scheme, a local user having read rights on the
472 dumpdates file could be able to do a Denial of Service attack
473 on dump. In order to lock the dumpdates file with the new
474 scheme, the user would need to have write access on the file.
475 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
476 reporting the bug (originally a bugtraq post).
478 2. Fixed interactive 'ls' which caused spurious errors warnings
479 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
480 <jorgen_ostling@users.sourceforge.net> for reporting this
483 3. Fixed dump's estimate when dealing with sparse inodes.
485 4. Modified dump to allow setting a blocksize bigger than 32kB
486 (raised the limit to 1024kB), because newer hardware needs
487 this for proper operation. Thanks to Dirk Traenapp
488 <dtraenapp@users.sourceforge.net> for reporting this.
490 5. Fixed a bug causing Dump to stop and report an error if an
491 inode number in the exclude file was followed by some amount
492 of whitespace. Thanks to Jeffrey Sofferin
493 <sofferin@users.sourceforge.net> for reporting this bug.
495 6. Fixed a bug which caused restore, in some particular cases,
496 to ask some 'scary' questions and leave a bunch of RSTTMP
497 directories behind when restoring incremental tapes. Thanks
498 to Philippe Troin <phil@fifi.org> for reporting this bug and
499 providing the test cases.
501 7. Changed the wording when inodes are excluded from dump:
502 replaced 'Added inode 7 to exclude list' with
503 'Excluding inode 7 (resize inode) from dump', as suggested
504 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
506 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
507 ===================================================================
509 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
510 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
511 <eros@lamel.bo.cnr.it> for reporting the bug.
513 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
514 ===================================================================
516 1. Made dump print out the ext2 logical block number in case of a read
517 error, which can be given as an argument to debugfs' ncheck command,
518 in order to find out the affected inode. Added note about this
519 usage in dump's man page.
521 2. Fixed a problem in restore when reading tapes written on big
522 endian machines with very old versions of dump. The patch was
523 contributed by George Helffrich <george@geology.bristol.ac.uk>.
525 3. Fixed the tape length calculation when using large tapes
526 and compression. Thanks to Georg Lippold
527 <g_lippold@sourceforge.net> for reporting the bug.
529 4. Added a new set of examples in dump_on_cd_2 directory, based
530 on dump_on_cd examples but somewhat enhanced, supporting DVD
531 media, and localized in english and german. Thanks to
532 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
534 5. Made dump save 32 bit UID/GID. Internally, this was achieved
535 by switching from the old BSD inode format to the new BSD
536 inode format, which means that the tape format was changed.
537 However, since all restore versions out there should
538 transparently support both inode formats, the change should
539 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
542 6. Fixed a lot of warnings in the code shown when compiling
543 with 'gcc -W'. Thanks to Matthias Andree
544 <matthias.andree@stud.uni-dortmund.de> for reporting this.
546 7. Fixed a small markup bug in the dump man page. Thanks to
547 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
550 8. Rewrote entirely the man pages using the tmac.an macro
551 package (Linux man page format) instead of the original BSD
552 format. They should be now cleaner and easier to modify.
554 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
555 ==================================================================
557 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
558 non Linux clients were misinterpreted. The description of the problem
559 (incompatible numbering in Linux mtio opcodes) is documented at
560 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
561 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
562 providing an excellent, cross-platform replacement for rmt in his
565 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
566 which caused restore to fail when the inode maps were not entirely
567 contained on the first volume (when using really small volumes or
568 when dumping a huge number of inodes).
570 3. Fixed a problem in dump, where files in subdirectories of directories
571 that have the nodump flag set (or being excluded with -e/-E) were
572 sometimes incorrectly dumped (instead of being excluded). The original
573 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
574 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
575 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
576 help testing the patch.
578 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
579 which makes one able to pipe the output of dump, by the net, to
580 a remote CD-burner server.
582 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
583 and use GNU's symbolic syntax over rmt instead of numerical values
584 to assure multiple platform compatibility.
586 6. Documented the -d option in restore.
588 7. Added a -v (verbose) mode to dump. For now it just prints the number
589 of the inode being dumped, but this could evolve in future versions
590 to include interesting debugging output.
592 8. Added a -o flag to restore, which automatically restores the current
593 directory permissions in -i or -x mode, without asking the operator
594 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
595 and Peter Samuel <peters@e-smith.com>.
597 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
598 ====================================================================
600 1. Fixed a bug in the QFA feature which could have caused
601 a bad tape positionning on restore (causing extra delays in
602 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
604 2. Added a small note in the dump man page specifying that
605 there should be no white space between the option letter and
606 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
608 3. Made restore work with older versions of the readline library,
609 by checking in configure for several extended symbols. Restore
610 can now be compiled with a readline as old as the 2.0 release
611 (though it may be a good idea to upgrade it to a more recent
612 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
613 reporting the build failures.
615 4. Fixed a performance problem with the QFA file creation in
616 dump, which made unnecessary seeks on the tape slowing down
617 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
618 reporting this issue.
620 5. Removed the inclusion of some kernel headers in the dump
621 source, which prevented the compile in some kernel/glibc
622 headers/architecture combination. Thanks to Bdale Garbee
623 <bdale@gag.com> for reporting the bug.
625 6. Added the appropriate error message when dump fails to
626 open the output file for writing. Thanks to Amith Varghese
627 <amithv@yahoo.com> for reporting this bug.
629 7. Made restore able to understand large Solaris ufsdump tapes
630 (containing inodes bigger than 4194304). Sun have introduced
631 an "extension" to the dump tape format when dealing with
632 those inodes, which was uncorrectly handled by Linux restore.
633 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
634 providing a test case.
636 8. Added the -m parameter to dump which optimises the output for
637 inodes having been changed but not modified since the last dump
638 ('changed' and 'modified' have the meaning defined in stat(2)).
639 For those inodes, dump will save only the metadata, instead of
640 saving the entire inode contents. Inodes which are either
641 directories or have been modified since the last dump are saved
642 in a regular way. Uses of this flag must be consistent, meaning
643 that either every dump in an incremental dump set have the flag,
646 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
647 =======================================================================
649 1. Fixed behaviour of dump when exceeding resource limits
652 2. Added the -L flag to restore to allow the user to specify a
653 maximal allowed number of miscompares when using restore
654 with the -C option to check the backup.
656 3. Detailed the manual entry for the -N option of restore.
658 4. Added the -a flag to restore to make able doing unattended
659 restores in -i or -x mode (automatically walks through the
660 multiple dump volumes).
662 5. Extended the QFA mode to work with local files and/or
663 remote tapes and files. This way, restore can know in advance
664 the tape number and the offset for the inodes to extract and
665 can minimize the extraction time by seeking directly to the
668 6. Added the -A <archive> option to both dump and restore,
669 which makes dump to archive a dump table-of-contents in
670 the specified file to be used by restore to determine
671 whether a file is in the dump file that is being restored.
672 (the archive file syntax is also compatible with the
673 Solaris ufsdump generated one).
675 7. Small fix in restore making it able to read some (broken ?)
676 Solaris ufsdump tapes.
678 8. Fixed dump to correctly recognise the root filesystem when using
679 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
680 for reporting this bug.
682 9. Added the -P <file> option to restore to create a
683 Quick File Access file from an already made dump. Patch
684 contributed by Uwe Gohlke <uwe@ugsoft.de>.
686 10. Made restore compile and run on Solaris, making it a
687 possible replacement for the standard ufsrestore. Port was
688 contributed by Uwe Gohlke <uwe@ugsoft.de>.
690 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
691 =====================================================================
693 1. Added a set of backup scripts from Eugenio Diaz
694 <getnito@yahoo.com> in the examples section. It features
695 automatic (cron based) full and incremental dumping of
696 several filesystems on a separate filesystem.
698 2. Fixed a off-by-one miscalculation which disabled dumping
699 a one letter subdirectory. Thanks to Chris Siebenmann
700 <cks@utcc.utoronto.ca> for reporting the bug.
702 3. Fixed several restore bugs occuring when trying to
703 restore some missing files on the tape. Thanks to Chris
704 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
706 4. Fixed --with-ldopts configure argument passing, installing from
707 a separate object directory, makefile cleanups contributed
708 by <splite@purdue.edu>.
710 5. Fix a bug which could caused, in some conditions, the highest
711 number inode of a filesystem, to not be dumped. Many thanks
712 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
715 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
716 =======================================================================
718 1. Added a mini howto from Patrick Walsh in the examples
721 2. Minor man pages syntax corrections. Thanks to
722 Chris Danis <screechco@home.com> for reporting the bugs.
724 3. Added a script from David B. Peterson <dave@toppledwagon.com>
725 to the examples section. It features dumping several
726 filesystems to a remote tape drive upon ssh.
728 4. Added a patch provided by Richard Jones <rich@annexia.org>
729 which allows BRADEMAX (number of read errors tolerated by
730 dump) to be adjusted using the -I option.
732 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
733 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
736 6. Add the -l option to restore to specify if, when doing a
737 remote restore, the file used to access the backup is a
738 regular file (the defaults being a tape drive). Restore needs
739 to know this information when reading a remote compressed
740 dump. Previously, this information was autodetected, but
741 the autodetection code fails (with ioctl: Inappropriate ioctl
742 for device) when using a non Linux remote box. Thanks to
743 many users and especially to Eros Albertazzi
744 <eros@lamel.bo.cnr.it> for reporting this.
746 7. Found a workaroung for the dump deadlock problem (3 childs
747 stuck in pause(), father in read()). The workaround seems
748 to work for me and several beta-testers. If it doesn't work
749 for you, please report back.
751 8. Updated the RPM spec file (BuildPrereq, URL etc).
753 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
754 ========================================================================
756 1. Fixed the permissions of a newly created QFA file by dump.
758 2. Cleaned up the source of dump (the external variables
759 definition was a complete mess, making possible to have
762 3. Fixed restore to use the full tape volume path when doing
763 a compare (since it changes the working directory to the
764 filesystem being compared in the process).
766 4. Added the -q option to dump which makes dump abort
767 whenever it needs operator attention. It should be
768 easier to use dump in scripts with this option.
770 5. Detect the use of incompatible options to dump and
771 refuse them (like -a and -B options together).
773 6. Added bzip2 compression to dump/restore (use option -j level
774 to select it). Note that you will need a restore version
775 >= 0.4b24 in order to restore a bzip2 compressed dump.
776 The same warning as for the zlib compression applies:
777 the tape format of a bzip2 dump is not compatible with the
778 original BSD tape format.
780 7. Fixed a overflow problem in dump corrupting the dump when
781 very large files were encountered. Thanks to Vladimir Nadvornik
784 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
785 kernel buffer/page cache before starting the dump, helping
786 a bit those who use dump on mounted filesystems. Thanks to
787 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
790 9. Updated the RPM spec file following the RedHat changes
791 (dynamically linked binaries now in /usr/sbin etc).
793 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
794 which enables restore to recognise multi volume compressed dumps
795 done on CD. Included his scripts for dump (which pipe the dump
796 output directly into cdrecord) and restore. There is now
797 possible to to backups to CD on the fly!
799 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
800 ===================================================================
802 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
803 Marcus Meissner (Caldera International Security Dept.).
805 2. Implement the Sun rmt extended protocol. Patch submitted
806 by Ian Gordon <iangordon@users.sourceforge.net>.
808 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
809 the linux kernel header. This ensures that dump always has the
810 latest version of this file. Patch submitted by
811 Andreas Dilger <adilger@turbolinux.com>.
813 4. Report any filesystem present in either /etc/fstab with a
814 positive passno or /etc/dumpdates in dump -w output.
815 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
817 5. Fixed the looping problem in dump introduced in the
820 6. Changed the -B option of dump to limit the size of
821 _compressed_ output per volume if compression is on.
822 Patch contributed by Helmut Jarausch
823 <jarausch@igpm.rwth-aachen.de>. Note however that, since
824 it is impossible to predict the size of the compressed
825 data before doing the compression, the -B limit is a bit
828 7. Fixed a bug in reading the operator typed file/tape path for
829 the next volume in restore.
831 8. Implemented a "-F script" option to restore which permits the
832 user to specify a script which will be launched at the
833 beginning of each tape, useful for automatic programming of
834 tape changers for example. See the restore man page for the
835 script parameters and return codes.
837 9. Small fix for the QFA routines provided by Uwe Gohlke
838 <uwe@ugsoft.de>, and some recommendations for QFA uses in
841 10. Fixed the multivolume restoring where making a mistake
842 to the 'Mount next tape' prompt caused several blocks to
845 11. Enhanced the -e option of dump to take as a parameter a
846 comma separated list of inode numbers.
848 12. Added the -E option to dump which specify a file containing
849 inode numbers to exclude from the dump.
851 13. Fixed the compressed multi-volume dump + restore.
853 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
854 ==================================================================
856 1. Made dump report the number of blocks dumped per volume.
857 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
859 2. Fix a bug in dump where waiting too much at the 'change volumes'
860 question caused the volume to be rewritten. Thanks to
861 Graham Reed <greed@users.sourceforge.net> for reporting the
862 bug and providing a patch.
864 3. Added a compression option to dump, contributed by Jerry
865 Peters <gapeters@worldnet.att.net>.
867 WARNING: the compression works only on files/pipes and
868 drives supporting variable block size.
870 WARNING2: tapes created using the compression feature are
871 incompatible with the standard dump tape format, and a
872 version of restore >= 0.4b22 will be needed for extraction.
874 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
875 architectures. Thanks to Paul Slootman <paul@debian.org> for
876 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
879 5. Many cleanups (CPP defines, const char warnings, check of
880 ext2fs COMPAT flags, time_t cleanups, added libext2 version
881 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
883 6. Made --prefix option work in configure. All the install path
884 are now based on the configure parameters.
886 7. Added the Quick File Access mode in dump/restore, contributed
887 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
888 a file tape position for each inode, and this file is used by
889 restore (if called with parameter Q and the filename)
890 to directly position the tape at the file restore is currently
891 working on. This saves hours when restoring single files from
892 large backups, saves the tapes and the drive's head. Use
893 --enable-qfa option of configure to compile in the QFA support.
895 8. Added the possibility to dump several files and directories
896 in a single invocation of dump. Thanks to Uwe Gohlke
897 <uwe@ugsoft.de> for implementing this option.
899 9. Fixed the dumping and restoring of files with holes
900 (files ending with holes didn't get dumped or restored
903 10. Fixed a socket descriptor leak in dump, which leaved opened
904 3 file descriptors per dump process (and there is one dump
907 11. Fixed dump large file system support, by explicit use of
908 open64/lseek64/etc functions (explicit use needed because
909 e2fsprogs libraries don't behave well when compiled with
910 FILE_OFFSET_BITS=64).
912 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
913 ======================================================================
915 1. Fixed some bugs in the dump -w|-W logic introduced by
916 the previous version. Thanks to Andreas Dilger
917 <adilger@turbolinux.com> for his help on this one.
919 2. Fixed again a compilation problem when using old e2fs
920 headers (filesystem label related). Thanks to many users
921 who reported this stupid error.
923 3. Fixed a build problem on old lib5 based systems dealing with
924 _PATH_MNTTAB being undefined. Thanks to John Adams
925 <johna@onevista.com> for reporting the bug.
927 4. Improved the error detection in restore when creating
928 the temporary files in TMPDIR. Now restore will corectly
929 report a 'no space left on device' error instead of
930 strange errors which could imply an archive corruption.
931 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
932 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
934 5. Added the throughput information in dump's progression
935 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
938 6. Use libext2fs's inode scanning routines, which are particularly
939 robust when dealing with errors (especially when having some
940 bad blocks on the filesystem). Many thanks to Colin
941 <colin@horizon.com> for reporting the bug, investigating
942 the issues, finding a workaround, writing the patch and
943 fully testing it... (of course, if this patch will break
944 anything, he is to blame for :-)).
946 7. Made dump and restore LFS aware. Dump can dump a filesystem
947 containing large files, generate a large file on output and
948 restore can restore them. This feature is not enabled by
949 default, use --enable-largefile option of configure to enable
950 it (you need to have a LFS aware glibc though). Thanks to
951 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
952 and to Theodore T'so <tytso@valinux.com> for his always
955 8. Made dump ask upon a tape write error if it should rewrite
956 the current volume (assume this is a bad tape) or if it should
957 assume an end-of-tape condition (useful for tape drives which
958 behaves badly at the end of the tape). Thanks to Andreas
959 Dilger <adilger@turbolinux.com> for the suggestion.
961 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
962 =======================================================================
964 1. Fixed a small compilation problem due to a change
965 in the definintion of the struct sigaction in
966 glibc 2.0 / libc5. Thanks to Gunther Schlegel
967 <schlegel@riege.de> for reporting the bug and to
968 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
971 2. Modified the label and uuid analysis in order to be
972 self-contained (independant of kernel/libc headers). This
973 should fix the compile with older kernel/libc headers and
974 will preserve the functionality. Thanks to Bernhard Erdmann
975 <bernhard.erdmann@gmx.de> for reporting the bug.
977 3. The 'exclude inode' option, if applied to a directory
978 inode, excludes now correctly all the leaves of this
979 directory. Thanks to John R. Dennison
980 <gerdesas@users.sourceforge.net> for reporting the bug.
982 4. Fixed the '-e' option to disable the possibility
983 to exclude the root inode (which causes the dumps to
984 be unreadable by restore). Prevented array overflow
985 when multiple -e options are used.
987 5. Fixed dump to correctly interpret a filesystem argument
988 which is a mountpoint and it is not an absolute path
989 (as specified in the fstab). Thanks to Bernhard R. Erdmann
990 <be@berdmann.de> for reporting the bug.
992 6. Made dump able to backup files larger than 2 GB. Note that
993 dump still doesn't cope with files larger than 4 GB.
995 7. Restore the real uid and gid before invoking an external
996 RSH program (big hole when dump or restore is suid root!).
998 8. Get the values for the default tape device and the location
999 of fstab file from the system headers. Thanks to
1000 Andreas Dilger <adilger@turbolinux.com> for the patch.
1002 9. Made dump -w|-W report all recognized filesystems
1003 present in either /etc/fstab or /etc/dumpdates, and present
1004 the list in the same order as in fstab file. Thanks
1005 to Andreas Dilger <adilger@turbolinux.com> for the patch.
1007 10. Made dump's -a (automatic end of tape detection) the
1008 default. Specifying one of -B, -c, -d or -s options will
1009 override the EOM detection. Thanks to Andreas Dilger
1010 <adilger@turbolinux.com> for the patch.
1012 11. Save the ext2 filesystem label into the dump volume label.
1013 Specifying a volume label on the command line (-L option)
1014 will override this feature. Thanks to Andreas Dilger
1015 <adilger@turbolinux.com> for the patch.
1017 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
1018 =====================================================================
1020 1. Fixed the signal handling in dump (which I broke in 0.4b17)
1021 which was causing several strange dump failures (dump
1022 hanged or segmentation faults etc).
1024 2. Specified the default blocksize in the dump man page.
1026 3. Changed two info messages of restore to be written on stdout
1027 instead of stderr in order to leave stderr only for errors
1028 or warnings. Thanks to Stephen Atwell
1029 <satwell@urbana.css.mot.com> for the suggestion.
1031 4. Corrected an off by one calculation which prevented
1032 dumping a directory having a 1 character name length.
1033 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
1034 for reporting the bug.
1036 5. Reinforce dump in order to allow only level 0 and no
1037 -u option when dumping a subdirectory, and document
1038 this in the man page. Thanks to Bernhard Erdmann
1039 <bernhard.erdmann@gmx.de> for reporting the bug.
1041 6. Fixed a small harmless bug in restore which caused
1042 in some conditions hard links to be restored several
1043 times (therefore generation some warning messages).
1044 Thanks to Kenneth Porter <shiva@well.com> for
1047 7. Updated the RPM spec file to the latest RedHat version,
1048 providing FHS packaging and other cosmetic changes.
1049 You will need RPM version >= 3.0.5 in order to build the RPM.
1051 8. Updated the configure script to check for libtermcap
1052 before checking for libreadline, because we need this
1053 library in order to compile the readline support.
1055 9. Made dump understand the LABEL= and UUID= notation
1056 both in /etc/fstab and on the command line. Note that
1057 you will need the /proc filesystem in order to use
1058 these notations. Thanks to Erik Troan <ewt@redhat.com>
1059 for providing the patch.
1061 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
1062 ===================================================================
1064 1. Fixed a potential buffer overflow in restore. Thanks
1065 to Stan Bubrouski <satan@fastdial.net> for reporting
1068 2. Fixed a readline-related bug which prevented
1069 'cat DUMPFILE | restore -i -f -' from working. Thanks
1070 to Charles Karney <karney@users.sourceforge.net>
1073 3. Changed a few "panic" into "exit", causing restore to
1074 be more stable against some attacks (like the last one
1075 reported on Bugtraq, although the last version of restore
1076 was not vulnerable - just dumped core). Thanks to
1077 Andreas Hasenack <andreas@conectiva.com.br> for reporting
1080 4. Removed the suid-root bit on dump and restore in the
1081 default build (and generated RPMs). It should be safer
1082 now. Users who need the suid-root capabilities in order
1083 to make network backups should read first the man page
1084 of dump and enable the suid bit manually.
1086 5. Added -ltermcap to the compile parameters for restore
1087 when using readline support, in order to make the compile
1088 process work on some systems (SuSE ?). Thanks to
1089 Patrik Schindler <poc@pocnet.net> for reporting the bug.
1091 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
1092 ====================================================================
1094 1. The -F script is called now *only* at the end of a tape,
1095 not at the end of the dump. Thanks to Christian Haul
1096 <haul@informatik.tu-darmstadt.de> for the bug report.
1098 Normally, the device name and the current volume number
1099 are passed on the command line of the script. However,
1100 if you want the old style script (with no arguments
1101 passed on the command line), you can enable it in
1102 configure with the --enable-oldstylefscript.
1104 2. Use posix signal handling to preserve dump functionality
1105 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
1108 3. Made the exit code of restore in case of a 'C'ompare
1109 command reflect the result of the compare. An exit status
1110 of 0 means the dump archive is correct, 1 means tape errors,
1111 2 means that some files were modified. Thanks to Kenneth Porter
1112 <shiva@well.com> for the suggestion.
1114 4. Made (finally) quotes work in interactive restore.
1116 5. Small fixes in order to allow dump to compile with a
1117 really old e2fsprogs version. Thanks to Ian Zimmerman
1118 <itz@speakeasy.org> for the bug report.
1120 6. Add GNU readline capabilities to interactive restore.
1121 Use configure's --enable-readline flag to enable this feature.
1122 Thanks to Patrik Schindler <poc@pocnet.net> for the
1125 7. Do the compare on the fly in restore 'C' mode (this will
1126 allow not to exhaust the available /tmp space when
1127 ccmparing large files). Thanks to Kenneth Porter
1128 <shiva@well.com> for the suggestion.
1130 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
1131 ====================================================================
1133 1. Fixed some several duplicate 'DUMP: DUMP:' in the
1136 2. Corrected the estimation of blocks to dump. Note that
1137 this is still wrong for large directory inodes, due
1138 to the size of a BSD directory entry on the tape
1139 (estimation is based on the size of the ext2 directory,
1140 which is smaller because it doesn't contain the null
1141 character at the end).
1143 3. Made dump report the total number of MB written to
1144 tape at the end of run. Thanks to W. Reilly Cooley
1145 <wcooley@nakedape.cc> for the patch.
1147 4. Added the -X option to restore giving the possibility
1148 to read the names of the files to be extracted/listed
1149 from a text file (in addition of the command line).
1150 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
1153 5. Added the device name and the current volume number
1154 as arguments to the end of tape script (dump -F option).
1156 6. Made the multi-volume dump work again (after having
1157 broken it in 0.4b15).
1159 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
1160 ===================================================================
1162 1. Added a prompt command in interactive restore mode. Thanks
1163 to Andreas Dilger <adilger@home.com> for the patch.
1165 2. Fixed a buffer overflow problem in dump (caused by
1166 not checking the size of the filesystem parameter).
1167 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
1168 reporting this on Bugtraq (and to several dump users
1169 who forwarded me his mail).
1171 3. Added the '-F script' option to dump in order to
1172 launch a script at the end of each tape (to be used
1173 with a tape changer, or to notify the sysadmin by
1176 4. Fixed a bug in restore compare code caused by the changes
1179 5. Fixed the treatment of options using the old BSD syntax
1180 in both dump and restore.
1182 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
1183 =======================================================================
1185 1. Fixed a bug in dump which may have caused invalid deleted
1186 directories to be dumped out if they were deleted after the
1187 mapping pass. This could occure on active filesystem and lead
1188 to heap corruption (causing dump malfunction in many possible ways).
1189 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
1190 bug and submitting the patch.
1192 2. Fixed the handling of the filesystem parameter in dump. A
1193 '/mnt/disk' parameter caused the disk contents to be dumped,
1194 but a '/mnt/disk/' parameter caused the mountpoint directory
1195 to be dumped (generally an empty directory).
1197 3. Improved the output of dump in order to tell which directory
1198 it is currently dumping (when dumping a subtree).
1200 4. Added the '-e' exclude inode option to dump. Thanks to
1201 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
1203 5. Added a REPORTING-BUGS file in order to provide a guide
1204 on how to correctly report a bug in dump/restore.
1206 6. Fixed a restore problem when restoring a hard link to an inode
1207 having the immutable or append-only attribute set. Thanks to
1208 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
1210 7. Fixed a compatibility problem between dumps made on little
1211 endian machines (the format was correct) and big endian
1212 machines (the format was incorrect). This fix break the
1213 compatibility with the older dumps made on big endian
1214 machines (sparc, m86k, ppc etc). For the first time in
1215 linux dump's history (I believe), the dumps made by *BSD,
1216 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
1217 so interchangeable. Thanks to Rob Cermak
1218 <cermak@ahab.rutgers.edu> for submitting the bug and
1219 helping me test the fix.
1221 8. Fixed the way dump reports the remaining percent/time, if
1222 the number of blocks actually dumped exceeds the estimated
1223 number of blocks. Thanks to Jean-Paul van der Jagt
1224 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
1226 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
1227 ======================================================================
1229 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
1230 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
1232 2. Removed the requirement to build the RPM as root from the
1233 spec file. Thanks to Christian Weisgerber
1234 <naddy@mips.rhein-neckar.de> for submitting this.
1236 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
1237 causing dump to block when dumping really huge filesystems.
1238 Many thanks to Patrik Schindler <poc@pocnet.net> for
1239 helping me find this bug.
1241 4. Fixed the treatment for an interrupt signal when dump access
1242 the remote tape through RSH. Thanks to Christian Weisgerber
1243 <naddy@mips.rhein-neckar.de> for providing the patch.
1245 5. Fixed a bug which was causing dump/restore to display
1246 garbage characters instead of the remote host name.
1248 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
1249 =====================================================================
1251 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
1252 <kukuk@suse.de> for submitting the patch.
1254 2. Fix for the exit code when using the size estimate option of
1255 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
1256 submitting the patch.
1258 3. Handle EINTR in atomical reads/writes in dump, which was causing
1259 dump to fail on some systems. Thanks to Eric Jergensen
1260 <eric@dvns.com> for reporting the bug and submitting the patch.
1262 4. Handle more than 16 characters for the device names in dumpdates.
1263 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
1264 tracking down the problem and proposing the solution.
1266 5. Fixed a bug in dump which prevented the creation of the
1267 dumpdates file when doing a 0-level dump without already
1268 having a dumpdates file. Thanks to Patrik Schindler
1269 <poc@pocnet.net> for reporting the bug.
1271 6. Changed the way dump 'S' flag reports the size estimate
1272 from number of blocks into bytes (making it compatible
1273 with the Solaris version, and simplifying things for
1274 amanda users). Thanks to Jason L Tibbitts III
1275 <tibbs@math.uh.edu> for reporting the bug.
1277 7. Fixed a compatibility problem in linux/alpha dump tape format.
1278 Now the linux/alpha dump are (again) compatible with the
1279 other dump formats. But this breaks compatibility with
1280 older dumps made on alpha. Thanks to Mike Tibor
1281 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
1283 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
1284 ======================================================================
1286 1. Added a '--enable-kerberos' to configure.
1288 2. Added a 'S' option to dump which determines the amount of space
1289 that is needed to perform the dump without actually doing it, similar
1290 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
1291 <cermak@ahab.rutgers.edu>.
1293 3. Added a 'M' multi-volume option to dump and restore which enables
1294 dumping to multiple files (useful when dumping to an ext2
1295 partition to make several dump files in order to bypass the 2GB
1296 file size limitation). The 'f' argument is treated as a prefix and
1297 the output files will be named <prefix>001, <prefix>002 etc. With
1298 the 'M' flag, restore automatically selects the right file without
1299 asking to enter a new tape each time.
1301 4. Fixed a memory leak which was causing dump to grow very big
1302 (270MB when dumping a 10GB filesystem...). Thanks to Jason
1303 Fearon <jasonf@netrider.org.au> for reporting the bug.
1305 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
1306 ======================================================================
1308 1. Make configure test if the system glob routines support
1309 extended syntax (ALTDIRFUNC). If not, use the internal glob
1310 routines instead of system ones. Thanks to Bernhard Sadlowski
1311 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
1312 and helping me resolve this and other minor libc5 compiling
1315 2. Fix a problem when dumping a ext2fs with the 'filetype'
1316 feature enabled. Thanks to Patrick J. LoPresti
1317 <patl@cag.lcs.mit.edu> for reporting the bug and to
1318 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
1320 3. Made the nodump flag work on directories. A directory which
1321 has the nodump flag gets never dumped, regardless of its
1324 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
1325 which allows dump on an active ext3 filesystem. However, this
1326 is a "quick and dirty" patch which enables backup of an ext3
1327 filesystem through the ext2 compatibility (by ignoring the
1328 NEEDS_RECOVERY bit). The journal file is not recognized and
1329 it is dumped (it should not).
1331 5. Test the superblock compatibility flags when dumping, in order
1332 to be sure that we know how to deal with specific features.
1334 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
1335 ====================================================================
1337 1. Use lchown instead of chown, fixing a possible security problem
1338 when restoring symlinks (a malicious user could use this
1339 to deliberately corrupt the ownership of important system files).
1340 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
1341 this and providing the patch.
1343 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
1344 ====================================================================
1346 1. Put dump sources under CVS, added Id tags in all files so
1347 one can use 'ident' on binary files.
1349 2. Added the dump/restore version in the usage text so one can
1350 easily verify the version he is using.
1352 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
1353 a va_start/va_end problem on linux-ppc (always call va_start
1354 va_end in pairs each time we use a vararg function).
1356 4. Added again the DT_* constants because old libc does not
1357 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
1358 for submitting the bug report.
1360 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
1361 this should enable dumping big (huge) filesystems.
1363 6. Added the RSH environment variable in order to be able to
1364 use a rsh replacement like ssh when doing remote backups (and
1365 bypass the security limitations of rcmd). Now you can do remote
1366 backups without being root (or making dump setuid root).
1368 7. Modified again the way dumpdates works. For incremental dumps,
1369 we need to read dumpdates even if we are not using 'u' option.
1370 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
1373 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
1374 ===================================================================
1376 1. Removed the 'k' flag from the restore 'about' text if kerberos
1377 was not compiled in.
1379 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
1380 to them (fsetflags takes a char*, setflags an open fd!).
1382 3. (f)setflags is called only if the flags aren't empty. If the
1383 file is a special file, a warning is printed, because changing
1384 flags implies opening the device. Normally, a special file
1385 should not have any flag... (Debian bug #29775, patch provided
1386 by Abhijit Dasgupta <abhijit@ans.net>).
1388 4. Made possible to dump a file system not mentioned in /etc/fstab.
1389 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
1391 5. Changed the default behaviour to not create dumpdates
1392 unless 'u' option is specified. Removed the old "debian-patch"
1393 which provided the same thing. (Debian bug #38136, #33818).
1395 6. Removed all those dump*announce, since they were getting old...
1397 7. Added warning messages if dumpdates does not exist and
1398 when an inferior level dump does not exist (except for a level 0
1401 8. Debugged the glob calls in interactive mode: restore used a
1402 dirent struct which was different from the /usr/include/dirent.h
1403 one (this used to work, is it a glibc2 change?), so none of the
1404 compat glob (which used /usr/include/dirent.h) or the system glob
1405 worked. Restore use now the system dirent (and the system
1406 DT_* constants), which are compatible with BSD ones.
1408 9. Added a configure flag (--with-dumpdatespath) to specify
1409 the location of dumpdates file. By default, it is
1412 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1413 included the current date/version in man pages.
1415 11. Corrected the estimation of remaining time when
1416 the operator doesn't change the tapes quickly enough. This
1417 was an old bug, I thought I corrected it, and discovered
1418 that in fact it was corrected in two different places, so
1419 the results canceled each other...
1421 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1422 ===================================================================
1424 1. Integrated multiple patches from RedHat, Debian and SuSE:
1426 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1428 - several fixes in the man pages.
1429 - update the default tape device to /dev/st0.
1430 - many updates for Linux Alpha (byte ordering, size_t etc).
1432 - use environment variable for TMPDIR (instead of /tmp).
1433 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1434 - workaround egcs bug (RedHat bugs #4281 and #2989).
1435 - wire $(OPT) throughout Makefile's.
1437 2. Upgrade the dump revision to 1, making possible to dump filesystems
1438 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1440 3. Fix some compile warnings, prototype all functions.
1442 4. Use glibc err/glob instead of internal compatibility
1443 routines (only if available).
1445 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1446 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1448 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1449 ======================================================================
1451 1. Integrated the changes from FreeBSD-3.1-RELEASE
1452 (mostly bug fixes, buffer overruns, dump has now an "automatic
1453 tape length calculation" flag, dump/restore can use kerberos now
1454 (this is NOT tested), use environment variables for TAPE and
1457 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1459 3. Corrected a bug in restore when using 'C' option with multi-volumes
1460 tapes (files splited accros two tapes give "size changed" errors
1463 4. Corrected the long standing bug when dumping multiple tapes.
1464 This works for me, needs further testing.
1466 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1467 ====================================================================
1469 1. Dump now runs correctly on kernels 2.1.x
1470 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1472 Changes between versions 0.4b2 and 0.4b3
1473 ========================================
1475 1. Use realpath() if available
1477 2. Report statistics
1479 Changes between versions 0.4b1 and 0.4b2
1480 ========================================
1482 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1485 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1487 3. Fixed configure.in to correctly handle --enable-rmt
1489 Changes between versions 0.3 and 0.4b1
1490 ======================================
1492 1. Integrated the changes from 4.4BSD-Lite2
1494 2. Integrated the patches from Debian and Red Hat
1496 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1498 4. Changed dump to use the Ext2fs library to get block addresses. This
1499 should solve the endianness problem on SparcLinux.
1501 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1502 distribution's one) to use autoconf
1504 6. Fixed a few minor bugs
1506 Changes between versions 0.2e and 0.2f
1507 ======================================
1509 1. Added the creation of named pipes (fifos) in restore.
1511 2. Added the -N flag in the restore manual page.
1513 3. Added the file kernel.patch which contains the llseek() optimization
1514 patch for 1.2.x kernels.
1516 4. Fixed a bug in the restoration of symbolic links: owner and group were
1519 5. Integrated some changes from FreeBSD 2.2.
1521 6. Added a call to ftruncate() after restoring each file to restore
1522 correctly files ending by a hole.
1524 Changes between versions 0.2d and 0.2e
1525 ======================================
1527 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1528 with owner = root (0) and group = root/wheel/whatever (0).
1530 Changes between versions 0.2c and 0.2d
1531 ======================================
1533 1. Dump is now able to backup 2GB+ filesystems.
1535 2. Dump and restore can now be linked as static binaries.
1537 Changes between versions 0.2b and 0.2c
1538 ======================================
1540 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1542 Changes between versions 0.2a and 0.2b
1543 ======================================
1545 1. Really fixed the bug that I should have corrected in 0.2a.
1547 2. Enabled optimization again.
1549 Changes between versions 0.2 and 0.2a
1550 =====================================
1552 1. Disabled the optimization during compilation.
1554 Changes between versions 0.1 and 0.2
1555 ====================================
1557 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1558 the fs_type field (actually, I modified the file fstab.c to make it
1559 use the mntent functions).
1561 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1562 made by dump should be readable by the BSD restore and backups made
1563 by the BSD dump should be readable by restore. Unfortunately, this
1564 means that the dump format has changed between version 0.1 and version
1567 3. Dump is now able to backup a subtree, it is no longer limited to whole
1568 filesystems like the BSD version.
1570 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1573 Changes between versions 0.0 and 0.1
1574 ====================================
1576 1. Now create links rdump and rrestore during the `make install' step.
1578 2. Linux port specific bugs added to the manual pages
1580 3. Incorrect estimation of the number of tapes blocks fixed when doing
1581 incremental backups.
1583 4. Better ls-like format in restore in interactive mode.
git.wh0rd.org / dump.git / blob