1 $Id: CHANGES,v 1.294 2008/06/09 13:25:40 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
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
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
36 8. Force '-a' to be the default value when running 'restore -C'.
37 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
40 9. Many improvements to the 'cron_dump_to_disk' example.
41 Thanks to Aaron S. Hawley <ashawley@users.sourceforge.net> for
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 Changes between versions 0.4b40 and 0.4b41 (released January 2, 2006)
54 =====================================================================
56 1. Fix restore of dumped Access Control Lists. The previous
57 code for EA works fine, but ACLs needed conversion from the
58 ext2/3 disk format to posix_acl format before restoring.
60 2. Fix some issues with restoration of EA on big endian
63 3. Fix restore when the symtab is over 2GB in size.
65 4. Made the directory hash indexing an optional feature,
66 accessible by the '-H' option of restore, and disable it
69 5. Fixed dump to not include extended attributes information
70 in the toc (archive) file which confused restore -t.
72 Changes between versions 0.4b39 and 0.4b40 (released May 2, 2005)
73 =================================================================
75 1. Changed restore to emit warnings (instead of emitting a fatal
76 error) if a file (or a directory) is unavailable for a
77 comparision (if the user doesn't have the necessary permissions
78 to access it for example). Thanks to Kenneth Porter
79 <shiva@sewingwitch.com> for the bug report.
81 2. Re-done the 'do not save directory entries to non-dumped inodes
82 (excluded from dump)' feature. The previous implementation
83 worked well for excluded directories but not for regular files.
84 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
87 3. Fixed a bug in dump where the tape size was miscalculated when
88 the user used -d/-s to specify the tape characteristics. Thanks
89 to Philip Goisman <goisman@physics.arizona.edu> for reporting
92 4. Fixed another bug introduced in restore with the hashtree
93 implementation. This one caused restore to stop saying
94 "removenode: non-empty directory" in some cases.
96 5. Added support for dumping and restoring ext2/3 extended
97 attributes (EA), like the access control lists (ACL) or
98 the security labels used by SELinux.
100 Changes between versions 0.4b38 and 0.4b39 (released January 21, 2005)
101 ======================================================================
103 1. The newly added dump_on_cd_3 example was buggy, replace it
104 with an updated version from Andrew Basterfield
105 <bob@cemetery.homeunix.org>.
107 2. Made restore to chdir() back into the initial directory when
108 dumping core while aborting a comparision operation. The
109 previous behaviour was to write the corefile at the root of
110 the directory being compared, which could very well be
111 read only and preventing the corefile generation. Thanks
112 to Kenneth Porter <shiva@sewingwitch.com> for the bug report.
114 3. Silenced the failure to call fgetflags() when comparing an
115 entry which has no ext2 attributes (as in lsattr()).
117 4. Fix a brown paper bug in restore -C which broke restore and
118 caused modifications on the filesystem being compared
119 (directories containing a file with the same name as the
120 directory get renamed to RSTTMP...). Thanks to Kenneth Porter
121 <shiva@sewingwitch.com> for finding the bug and helping me
124 5. Made restore -C force the -N flag (no writing allowed on
125 the disk) in order to prevent more bugs like the above one.
127 Changes between versions 0.4b37 and 0.4b38 (released January 7, 2005)
128 =====================================================================
130 1. Fix a couple of troff syntax bugs in the man pages.
131 Thanks to Eric Raymond <esr@thyrsus.com> for the patch.
133 2. Made restore use either libncurses or libtermcap, depending
134 on which one is available at configure time.
136 3. Fixed restore negative size display bug when comparing a
137 dump containing files over 2GB. Thanks to Steve Bonds
138 <sbonds@users.sourceforge.net> for the bug report.
140 4. Do not save directory entries to non-dumped inodes
141 (excluded from dump). This will eliminate the 'missing
142 file' warnings when doing 'restore -C'.
144 5. Fix dump crash when backuping a huge (2TB) filesystem,
145 due to a bogus calculation on the inode map size.
146 Thanks to Kevin B. Haines <K.B.Haines@rl.ac.uk> for
147 submitting the bug and testing the fix.
149 6. Fix a problem in restore where the final \0 in the symbolic
150 link names could have been lost, generating corrupt filenames.
151 Thanks to Kyle Wilson <kyle.wilson@amd.com> for reporting the
154 7. Implemented a hash list for the directory names in restore.
155 The linear list used before caused problems in interactive
156 restores when dealing with directories having thousands of
157 entries. Thanks to Brian Ristuccia <bristuccia@starentnetworks.com>
158 for reporting the bug.
160 8. Improved restore -C, this time including the directory
161 attributes into the comparision.
163 9. Made restore understand tapes containing EA/ACLs (which will
164 be dumped by the next version of dump). In this version
165 extended attributes on the tape are ignored, for full EA/ACL
166 support wait for the next version or try the experimental EA
169 Changes between versions 0.4b36 and 0.4b37 (released July 7, 2004)
170 ==================================================================
172 1. Added the --enable-staticz configure option which enables
173 dump and restore to be linked with static versions of
174 libz and libbz2 (and dynamic versions of all other libraries).
175 This will make Debian users happy, because libz and libbz2
176 were the only needed libraries living in /usr, all the
177 others live in /lib. In case of system emergency, it is
178 better not to have to rely on an extra filesystem.
179 Thanks to Bdale Garbee <bdale@gag.com> for the suggestion.
181 2. Fix compilation on (at least the Linux Debian port to) AMD64.
182 (<ext2fs/ext2_types.h> defines some types (__s64 and __u64)
183 that are also defined by <linux/types.h> (<asm/types.h>) and
186 3. Make dump's reading of the dumpdates file a bit more robust,
187 preventing dump from crashing when the dumpdates file has
188 been modified by hand.
190 4. Fixed some offset calculations in dump code which could
191 lead to "bread lseek errors" on large filesystems. Thanks
192 to Bruce Lowekamp <lowekamp@users.sourceforge.net> for
193 reporting this bug and debugging the issue.
195 5. Made dump use the blkid library when searching for devices
196 by label or uuid instead of dump's own routines.
198 6. Corrected a bug in dump where a wrong LABEL=... line in
199 /etc/fstab could prevent dump from dumping unrelated
200 filesystems. Thanks to Bruce Lowekamp
201 <lowekamp@users.sourceforge.net> for reporting the bug.
203 Changes between versions 0.4b35 and 0.4b36 (released April 21, 2004)
204 ====================================================================
206 1. Fixed dump compilation with old gcc versions. Thanks to
207 Mike Castle <dalgoda@users.sourceforge.net> for the patch.
209 2. Fixed some warnings (howmany, roundup, powerof2 redefined)
210 when compiling against a recent glibc version.
212 3. Fixed a bug in restore preventing the read of a dump tape
213 written with Solaris 7 ufsdump. Thanks to Patrick Higgins
214 <phiggins@transzap.com> for reporting the bug and providing
217 4. Changed dump to enable the creation of volumes of different
218 sizes in a single run (make -B accept a list of values).
219 Patch contributed by Florian Zumbiehl <florz@gmx.de>.
221 5. Use the glibc provided minor() and major() macros instead
222 of our own bitmask implementation. This should be safe for
223 when the major/minor namespace will migrate to 32 bits.
224 Thanks to Zhang Jun <zhangjun@nanjing-fnst.com> for reporting
227 6. Made explicit in the dump man page that dump will not create
228 a remote file, it will only write to an already existing one.
230 7. Another try at making size estimates better again.
232 8. Put back the inconditional running of the end-of-tape
233 script which was changed in 0.434 to be run only when -M
234 or multiple -f were NOT specified. Some users rely on this
235 feature even when it is combined with -M/-f.
237 9. Fixed restore when restoring huge backups (where rstdir...
238 temporary files are over 2GB). Thanks to
239 Raphael Jaffey <rjaffey@artic.edu> for reporting this,
240 debugging the issue and making the patch.
242 10. Made restore understand FreeBSD UFS2 tapes. Thanks to
243 David <vrtin@users.sourceforge.net> for submitting the bug
244 and providing a test case.
246 11. Made dump work with an arbitrary integer level (no more 10
247 levels only limitation). Thanks to Matthew
248 <msvincen@midway.uchicago.edu> for the patch.
250 Changes between versions 0.4b34 and 0.4b35 (released December 21, 2003)
251 =======================================================================
253 1. Added a note in the dump man page saying that the default
254 blocksize can be 32 if -d is used with a high density argument.
255 Thanks to Antonios Christofides <A.Christofides@itia.ntua.gr>
258 2. Fixed configure to correctly understand CPPFLAGS, CFLAGS,
259 LDFLAGS environment variables. Thanks to Arcady Genkin
260 <antipode@users.sourceforge.net> for reporting the bug.
262 3. Made -e/-E options of dump accept an unlimited number of inodes
263 to be excluded, rather than a hardcoded maximum. Thanks to
264 Dietrich Rothe <d-rothe@users.sourceforge.net> for the patch.
266 4. Updated the autoconf system to 2.50. Forced the -D_BSD_SOURCE
267 and -D_USE_BSD_SIGNAL defines in configure in order to solve
268 64bit build problems because quad_t is redifined with a
269 different signature. Thanks to Mike Harris <mharris@redhat.com>
270 for reporting this bug.
272 5. Made restore build on Solaris, making possible to
273 restore Linux's "enhanced" tapes. Thanks to Uwe Gohlke
274 <uwe@ugsoft.de> for the patch.
276 6. Made an extension in the dump tape format capable of saving
277 MacOSX specific inode extensions. Uwe Gohlke <uwe@ugsoft.de>
278 wrote the extension and contributed the restore code back
279 into this codebase. The same extension mechanism will be
280 used in the future to save ACLs...
282 7. Made rmt work correctly with regard to QFA and local/remote
283 files and tapes. The remote access will however work only
284 when the dump provided rmt version is used. If you want to
285 use another rmt server, please do not use the QFA feature.
286 Thanks to Uwe Gohlke <uwe@ugsoft.de> for the patch.
288 Changes between versions 0.4b33 and 0.4b34 (released April 18, 2003)
289 ====================================================================
291 1. Fixed the INSTALL file to reflect the actual install paths.
292 Thanks to David Ronis <ronis@ronispc.chem.mcgill.ca> for
295 2. Fixed the configure script to only check for headers presence
296 instead of trying to compile them. This should fix issues
297 with old build environments. Thanks to Kari Lempiainen
298 <kari@funky.pp.fi> for reporting the bug.
300 3. Fixed restore to correctly ignore sockets when comparing
301 a dump (as socket cannot be properly restored anyway). Thanks
302 to Gunther Reiszig <gunther@mit.edu> for reporting the bug.
304 4. Fixed restore to correctly access the archive file (-A argument)
305 even when using a remote tape. Thanks to Fabrice Bellet
306 <fabrice@bellet.info> for reporting the bug.
308 5. Fixed (again) handling of long (largefile) seeks in rmt.
309 Thanks to Fabrice Bellet <fabrice@bellet.info> for reporting
312 6. Fixed restore corner case when dealing with large block sizes
313 dump is able to write now (-b 1024). Thanks to Fabrice Bellet
314 <fabrice@bellet.info> for reporting the bug.
316 7. Fixed a bug preventing dump to access a filesystem having
317 a label exactly 16 bytes in length. Thanks to <minduni@ti-edu.ch>
318 for reporting the bug.
320 8. Made dump store dump dates using explicit timezones, fixing a
321 problem with incremental dumps if the timezone is modified
322 between the dumps. Thanks to Philippe Troin <phil@fifi.org> for
323 the bug report and the patch.
325 9. Fixed a bug encountered when dumping individual files (not full
326 filesystems or directories) and dangling symbolic links happen
327 to be in the list of files. For as far as dump is concerned,
328 dangling symbolic links are allowed, and are dumped as is.
329 Thanks to Jin-su Ahn <jsahn@ee.snu.ac.kr> for reporting the
330 bug and providing the fix.
332 10. Fixed open and creation modes and permissions for QFA and
333 table-of-contents files in dump and restore. Thanks to
334 Philippe Troin <phil@fifi.org> for the patch.
336 11. Fixed the archive file descriptor handling enabling it to be 0.
337 This can happen in some cases when shell redirections are used.
338 Thanks to Philippe Troin <phil@fifi.org> for the patch.
340 12. Delayed the opening of archive file until after suid had been
341 dropped (fixing a possible security issue if dump is suid).
342 Thanks to Philippe Troin <phil@fifi.org> for the patch.
344 13. Fixed the 'S' command handling in the rmt client part.
345 Thanks to Philippe Troin <phil@fifi.org> for the patch.
347 14. Modified the end-of-tape script handling to print out statistics
348 (and stop the timer) before launching the eot script. Also, the eot
349 script does not get run anymore when using -M (which makes sense) or
350 when multiple tapes are listed on the command line
351 (-f tape0,tape1,tapen) (which also makes sense).
352 Thanks to Philippe Troin <phil@fifi.org> for the patch.
354 15. Relicensed dump/restore under the 'revised' BSD license, as per
355 ftp://ftp.cs.berkeley.edu/ucb/4bsd/README.Impt.License.Change.
357 16. Added LZO compression to dump. This new compression method has
358 the advantage of being super fast, thus not killing tape streaming
359 on slow machines. Thanks to Helmut Jarausch
360 <jarausch@igpm.rwth-aachen.de> for the patch and to
361 Markus Oberhumer <markus@oberhumer.com> for giving special permission
362 to include his miniLZO project (GPL licensed) in dump/restore.
364 17. Some small buffer overruns fixes in rmt. Thanks to Antonomasia
365 <ant@notatla.demon.co.uk> for reporting the bugs.
367 18. Added a special rmt version which can do encryption when writing
368 to tape. Read examples/encrypted_rmt/README for details on
369 how to enable and configure it. Thanks to Ken Lalonde
370 <ken@globalremit.com> for the patch.
372 19. Made dump work with 2.5 kernel end of tape early warning semantics.
373 Thanks to Kai Makisara <Kai.Makisara@kolumbus.fi> for the patch.
375 20. Fixed a bug which caused dump -w|-W not to work anymore, because
376 the fs_freq and fs_passno fields in /etc/mtab are always set
377 to 0 0. Thanks to Trent Piepho <xyzzy@speakeasy.org> for
380 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
381 =======================================================================
383 1. Added a note in the restore man page clarifying the question
384 'set the permissions on the current directory ?' asked by
385 restore at the end of treatment in -i and -x modes.
387 2. Fixed the endianess issues when reading compressed tapes.
388 Thanks to Dark Force <daq4th@users.sourceforge.net> for
389 reporting this bug and providing test cases.
391 3. Fixed the "ACL won't be dumped" warning message (which showed
392 an extra, unrelated error message). Thanks to Dragan Krnic
393 <dkrnic@lycos.com> for reporting this bug.
395 4. Made dump look first into /etc/mtab, then into /etc/fstab
396 when searching for filesystem paths. Also fixed some problems
397 caused by binding mounts or multiple block device mounts.
398 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
399 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
400 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
401 and #160305, Sourceforge bugs #618699 and #679832).
403 5. Made dump's -I option accept the value '0' meaning all the
404 read errors will be ignored. This can be useful when running
405 dump from unattended sessions (like cron jobs). Thanks to
406 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
408 6. Fixed the output of dump to indicate 'blocks' instead of
409 'tape blocks' in the various messages (blocks are always
410 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
411 made it clearly print the current blocksize at the start of
412 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
415 7. Made rmt understand long (largefiles) seeks.
417 8. Fixed build with very old versions of libext2fs, where
418 EXT2_FT_* constants were undefined.
420 9. Made dump accept the dumpdates path on the command line
421 (-D file option) instead of using only the hardcoded one.
422 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
425 10. Enabled rmt, LFS, readline, QFA options by default in
426 ./configure. Updated the configure process (new versions
427 of config.guess, config.sub etc).
429 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
430 =======================================================================
432 1. Changed dump to use fcntl(F_SETLK) style locking instead
433 of flock() when locking the dumpdates file. With the old
434 locking scheme, a local user having read rights on the
435 dumpdates file could be able to do a Denial of Service attack
436 on dump. In order to lock the dumpdates file with the new
437 scheme, the user would need to have write access on the file.
438 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
439 reporting the bug (originally a bugtraq post).
441 2. Fixed interactive 'ls' which caused spurious errors warnings
442 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
443 <jorgen_ostling@users.sourceforge.net> for reporting this
446 3. Fixed dump's estimate when dealing with sparse inodes.
448 4. Modified dump to allow setting a blocksize bigger than 32kB
449 (raised the limit to 1024kB), because newer hardware needs
450 this for proper operation. Thanks to Dirk Traenapp
451 <dtraenapp@users.sourceforge.net> for reporting this.
453 5. Fixed a bug causing Dump to stop and report an error if an
454 inode number in the exclude file was followed by some amount
455 of whitespace. Thanks to Jeffrey Sofferin
456 <sofferin@users.sourceforge.net> for reporting this bug.
458 6. Fixed a bug which caused restore, in some particular cases,
459 to ask some 'scary' questions and leave a bunch of RSTTMP
460 directories behind when restoring incremental tapes. Thanks
461 to Philippe Troin <phil@fifi.org> for reporting this bug and
462 providing the test cases.
464 7. Changed the wording when inodes are excluded from dump:
465 replaced 'Added inode 7 to exclude list' with
466 'Excluding inode 7 (resize inode) from dump', as suggested
467 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
469 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
470 ===================================================================
472 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
473 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
474 <eros@lamel.bo.cnr.it> for reporting the bug.
476 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
477 ===================================================================
479 1. Made dump print out the ext2 logical block number in case of a read
480 error, which can be given as an argument to debugfs' ncheck command,
481 in order to find out the affected inode. Added note about this
482 usage in dump's man page.
484 2. Fixed a problem in restore when reading tapes written on big
485 endian machines with very old versions of dump. The patch was
486 contributed by George Helffrich <george@geology.bristol.ac.uk>.
488 3. Fixed the tape length calculation when using large tapes
489 and compression. Thanks to Georg Lippold
490 <g_lippold@sourceforge.net> for reporting the bug.
492 4. Added a new set of examples in dump_on_cd_2 directory, based
493 on dump_on_cd examples but somewhat enhanced, supporting DVD
494 media, and localized in english and german. Thanks to
495 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
497 5. Made dump save 32 bit UID/GID. Internally, this was achieved
498 by switching from the old BSD inode format to the new BSD
499 inode format, which means that the tape format was changed.
500 However, since all restore versions out there should
501 transparently support both inode formats, the change should
502 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
505 6. Fixed a lot of warnings in the code shown when compiling
506 with 'gcc -W'. Thanks to Matthias Andree
507 <matthias.andree@stud.uni-dortmund.de> for reporting this.
509 7. Fixed a small markup bug in the dump man page. Thanks to
510 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
513 8. Rewrote entirely the man pages using the tmac.an macro
514 package (Linux man page format) instead of the original BSD
515 format. They should be now cleaner and easier to modify.
517 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
518 ==================================================================
520 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
521 non Linux clients were misinterpreted. The description of the problem
522 (incompatible numbering in Linux mtio opcodes) is documented at
523 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
524 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
525 providing an excellent, cross-platform replacement for rmt in his
528 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
529 which caused restore to fail when the inode maps were not entirely
530 contained on the first volume (when using really small volumes or
531 when dumping a huge number of inodes).
533 3. Fixed a problem in dump, where files in subdirectories of directories
534 that have the nodump flag set (or being excluded with -e/-E) were
535 sometimes incorrectly dumped (instead of being excluded). The original
536 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
537 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
538 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
539 help testing the patch.
541 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
542 which makes one able to pipe the output of dump, by the net, to
543 a remote CD-burner server.
545 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
546 and use GNU's symbolic syntax over rmt instead of numerical values
547 to assure multiple platform compatibility.
549 6. Documented the -d option in restore.
551 7. Added a -v (verbose) mode to dump. For now it just prints the number
552 of the inode being dumped, but this could evolve in future versions
553 to include interesting debugging output.
555 8. Added a -o flag to restore, which automatically restores the current
556 directory permissions in -i or -x mode, without asking the operator
557 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
558 and Peter Samuel <peters@e-smith.com>.
560 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
561 ====================================================================
563 1. Fixed a bug in the QFA feature which could have caused
564 a bad tape positionning on restore (causing extra delays in
565 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
567 2. Added a small note in the dump man page specifying that
568 there should be no white space between the option letter and
569 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
571 3. Made restore work with older versions of the readline library,
572 by checking in configure for several extended symbols. Restore
573 can now be compiled with a readline as old as the 2.0 release
574 (though it may be a good idea to upgrade it to a more recent
575 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
576 reporting the build failures.
578 4. Fixed a performance problem with the QFA file creation in
579 dump, which made unnecessary seeks on the tape slowing down
580 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
581 reporting this issue.
583 5. Removed the inclusion of some kernel headers in the dump
584 source, which prevented the compile in some kernel/glibc
585 headers/architecture combination. Thanks to Bdale Garbee
586 <bdale@gag.com> for reporting the bug.
588 6. Added the appropriate error message when dump fails to
589 open the output file for writing. Thanks to Amith Varghese
590 <amithv@yahoo.com> for reporting this bug.
592 7. Made restore able to understand large Solaris ufsdump tapes
593 (containing inodes bigger than 4194304). Sun have introduced
594 an "extension" to the dump tape format when dealing with
595 those inodes, which was uncorrectly handled by Linux restore.
596 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
597 providing a test case.
599 8. Added the -m parameter to dump which optimises the output for
600 inodes having been changed but not modified since the last dump
601 ('changed' and 'modified' have the meaning defined in stat(2)).
602 For those inodes, dump will save only the metadata, instead of
603 saving the entire inode contents. Inodes which are either
604 directories or have been modified since the last dump are saved
605 in a regular way. Uses of this flag must be consistent, meaning
606 that either every dump in an incremental dump set have the flag,
609 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
610 =======================================================================
612 1. Fixed behaviour of dump when exceeding resource limits
615 2. Added the -L flag to restore to allow the user to specify a
616 maximal allowed number of miscompares when using restore
617 with the -C option to check the backup.
619 3. Detailed the manual entry for the -N option of restore.
621 4. Added the -a flag to restore to make able doing unattended
622 restores in -i or -x mode (automatically walks through the
623 multiple dump volumes).
625 5. Extended the QFA mode to work with local files and/or
626 remote tapes and files. This way, restore can know in advance
627 the tape number and the offset for the inodes to extract and
628 can minimize the extraction time by seeking directly to the
631 6. Added the -A <archive> option to both dump and restore,
632 which makes dump to archive a dump table-of-contents in
633 the specified file to be used by restore to determine
634 whether a file is in the dump file that is being restored.
635 (the archive file syntax is also compatible with the
636 Solaris ufsdump generated one).
638 7. Small fix in restore making it able to read some (broken ?)
639 Solaris ufsdump tapes.
641 8. Fixed dump to correctly recognise the root filesystem when using
642 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
643 for reporting this bug.
645 9. Added the -P <file> option to restore to create a
646 Quick File Access file from an already made dump. Patch
647 contributed by Uwe Gohlke <uwe@ugsoft.de>.
649 10. Made restore compile and run on Solaris, making it a
650 possible replacement for the standard ufsrestore. Port was
651 contributed by Uwe Gohlke <uwe@ugsoft.de>.
653 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
654 =====================================================================
656 1. Added a set of backup scripts from Eugenio Diaz
657 <getnito@yahoo.com> in the examples section. It features
658 automatic (cron based) full and incremental dumping of
659 several filesystems on a separate filesystem.
661 2. Fixed a off-by-one miscalculation which disabled dumping
662 a one letter subdirectory. Thanks to Chris Siebenmann
663 <cks@utcc.utoronto.ca> for reporting the bug.
665 3. Fixed several restore bugs occuring when trying to
666 restore some missing files on the tape. Thanks to Chris
667 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
669 4. Fixed --with-ldopts configure argument passing, installing from
670 a separate object directory, makefile cleanups contributed
671 by <splite@purdue.edu>.
673 5. Fix a bug which could caused, in some conditions, the highest
674 number inode of a filesystem, to not be dumped. Many thanks
675 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
678 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
679 =======================================================================
681 1. Added a mini howto from Patrick Walsh in the examples
684 2. Minor man pages syntax corrections. Thanks to
685 Chris Danis <screechco@home.com> for reporting the bugs.
687 3. Added a script from David B. Peterson <dave@toppledwagon.com>
688 to the examples section. It features dumping several
689 filesystems to a remote tape drive upon ssh.
691 4. Added a patch provided by Richard Jones <rich@annexia.org>
692 which allows BRADEMAX (number of read errors tolerated by
693 dump) to be adjusted using the -I option.
695 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
696 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
699 6. Add the -l option to restore to specify if, when doing a
700 remote restore, the file used to access the backup is a
701 regular file (the defaults being a tape drive). Restore needs
702 to know this information when reading a remote compressed
703 dump. Previously, this information was autodetected, but
704 the autodetection code fails (with ioctl: Inappropriate ioctl
705 for device) when using a non Linux remote box. Thanks to
706 many users and especially to Eros Albertazzi
707 <eros@lamel.bo.cnr.it> for reporting this.
709 7. Found a workaroung for the dump deadlock problem (3 childs
710 stuck in pause(), father in read()). The workaround seems
711 to work for me and several beta-testers. If it doesn't work
712 for you, please report back.
714 8. Updated the RPM spec file (BuildPrereq, URL etc).
716 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
717 ========================================================================
719 1. Fixed the permissions of a newly created QFA file by dump.
721 2. Cleaned up the source of dump (the external variables
722 definition was a complete mess, making possible to have
725 3. Fixed restore to use the full tape volume path when doing
726 a compare (since it changes the working directory to the
727 filesystem being compared in the process).
729 4. Added the -q option to dump which makes dump abort
730 whenever it needs operator attention. It should be
731 easier to use dump in scripts with this option.
733 5. Detect the use of incompatible options to dump and
734 refuse them (like -a and -B options together).
736 6. Added bzip2 compression to dump/restore (use option -j level
737 to select it). Note that you will need a restore version
738 >= 0.4b24 in order to restore a bzip2 compressed dump.
739 The same warning as for the zlib compression applies:
740 the tape format of a bzip2 dump is not compatible with the
741 original BSD tape format.
743 7. Fixed a overflow problem in dump corrupting the dump when
744 very large files were encountered. Thanks to Vladimir Nadvornik
747 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
748 kernel buffer/page cache before starting the dump, helping
749 a bit those who use dump on mounted filesystems. Thanks to
750 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
753 9. Updated the RPM spec file following the RedHat changes
754 (dynamically linked binaries now in /usr/sbin etc).
756 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
757 which enables restore to recognise multi volume compressed dumps
758 done on CD. Included his scripts for dump (which pipe the dump
759 output directly into cdrecord) and restore. There is now
760 possible to to backups to CD on the fly!
762 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
763 ===================================================================
765 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
766 Marcus Meissner (Caldera International Security Dept.).
768 2. Implement the Sun rmt extended protocol. Patch submitted
769 by Ian Gordon <iangordon@users.sourceforge.net>.
771 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
772 the linux kernel header. This ensures that dump always has the
773 latest version of this file. Patch submitted by
774 Andreas Dilger <adilger@turbolinux.com>.
776 4. Report any filesystem present in either /etc/fstab with a
777 positive passno or /etc/dumpdates in dump -w output.
778 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
780 5. Fixed the looping problem in dump introduced in the
783 6. Changed the -B option of dump to limit the size of
784 _compressed_ output per volume if compression is on.
785 Patch contributed by Helmut Jarausch
786 <jarausch@igpm.rwth-aachen.de>. Note however that, since
787 it is impossible to predict the size of the compressed
788 data before doing the compression, the -B limit is a bit
791 7. Fixed a bug in reading the operator typed file/tape path for
792 the next volume in restore.
794 8. Implemented a "-F script" option to restore which permits the
795 user to specify a script which will be launched at the
796 beginning of each tape, useful for automatic programming of
797 tape changers for example. See the restore man page for the
798 script parameters and return codes.
800 9. Small fix for the QFA routines provided by Uwe Gohlke
801 <uwe@ugsoft.de>, and some recommendations for QFA uses in
804 10. Fixed the multivolume restoring where making a mistake
805 to the 'Mount next tape' prompt caused several blocks to
808 11. Enhanced the -e option of dump to take as a parameter a
809 comma separated list of inode numbers.
811 12. Added the -E option to dump which specify a file containing
812 inode numbers to exclude from the dump.
814 13. Fixed the compressed multi-volume dump + restore.
816 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
817 ==================================================================
819 1. Made dump report the number of blocks dumped per volume.
820 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
822 2. Fix a bug in dump where waiting too much at the 'change volumes'
823 question caused the volume to be rewritten. Thanks to
824 Graham Reed <greed@users.sourceforge.net> for reporting the
825 bug and providing a patch.
827 3. Added a compression option to dump, contributed by Jerry
828 Peters <gapeters@worldnet.att.net>.
830 WARNING: the compression works only on files/pipes and
831 drives supporting variable block size.
833 WARNING2: tapes created using the compression feature are
834 incompatible with the standard dump tape format, and a
835 version of restore >= 0.4b22 will be needed for extraction.
837 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
838 architectures. Thanks to Paul Slootman <paul@debian.org> for
839 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
842 5. Many cleanups (CPP defines, const char warnings, check of
843 ext2fs COMPAT flags, time_t cleanups, added libext2 version
844 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
846 6. Made --prefix option work in configure. All the install path
847 are now based on the configure parameters.
849 7. Added the Quick File Access mode in dump/restore, contributed
850 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
851 a file tape position for each inode, and this file is used by
852 restore (if called with parameter Q and the filename)
853 to directly position the tape at the file restore is currently
854 working on. This saves hours when restoring single files from
855 large backups, saves the tapes and the drive's head. Use
856 --enable-qfa option of configure to compile in the QFA support.
858 8. Added the possibility to dump several files and directories
859 in a single invocation of dump. Thanks to Uwe Gohlke
860 <uwe@ugsoft.de> for implementing this option.
862 9. Fixed the dumping and restoring of files with holes
863 (files ending with holes didn't get dumped or restored
866 10. Fixed a socket descriptor leak in dump, which leaved opened
867 3 file descriptors per dump process (and there is one dump
870 11. Fixed dump large file system support, by explicit use of
871 open64/lseek64/etc functions (explicit use needed because
872 e2fsprogs libraries don't behave well when compiled with
873 FILE_OFFSET_BITS=64).
875 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
876 ======================================================================
878 1. Fixed some bugs in the dump -w|-W logic introduced by
879 the previous version. Thanks to Andreas Dilger
880 <adilger@turbolinux.com> for his help on this one.
882 2. Fixed again a compilation problem when using old e2fs
883 headers (filesystem label related). Thanks to many users
884 who reported this stupid error.
886 3. Fixed a build problem on old lib5 based systems dealing with
887 _PATH_MNTTAB being undefined. Thanks to John Adams
888 <johna@onevista.com> for reporting the bug.
890 4. Improved the error detection in restore when creating
891 the temporary files in TMPDIR. Now restore will corectly
892 report a 'no space left on device' error instead of
893 strange errors which could imply an archive corruption.
894 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
895 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
897 5. Added the throughput information in dump's progression
898 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
901 6. Use libext2fs's inode scanning routines, which are particularly
902 robust when dealing with errors (especially when having some
903 bad blocks on the filesystem). Many thanks to Colin
904 <colin@horizon.com> for reporting the bug, investigating
905 the issues, finding a workaround, writing the patch and
906 fully testing it... (of course, if this patch will break
907 anything, he is to blame for :-)).
909 7. Made dump and restore LFS aware. Dump can dump a filesystem
910 containing large files, generate a large file on output and
911 restore can restore them. This feature is not enabled by
912 default, use --enable-largefile option of configure to enable
913 it (you need to have a LFS aware glibc though). Thanks to
914 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
915 and to Theodore T'so <tytso@valinux.com> for his always
918 8. Made dump ask upon a tape write error if it should rewrite
919 the current volume (assume this is a bad tape) or if it should
920 assume an end-of-tape condition (useful for tape drives which
921 behaves badly at the end of the tape). Thanks to Andreas
922 Dilger <adilger@turbolinux.com> for the suggestion.
924 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
925 =======================================================================
927 1. Fixed a small compilation problem due to a change
928 in the definintion of the struct sigaction in
929 glibc 2.0 / libc5. Thanks to Gunther Schlegel
930 <schlegel@riege.de> for reporting the bug and to
931 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
934 2. Modified the label and uuid analysis in order to be
935 self-contained (independant of kernel/libc headers). This
936 should fix the compile with older kernel/libc headers and
937 will preserve the functionality. Thanks to Bernhard Erdmann
938 <bernhard.erdmann@gmx.de> for reporting the bug.
940 3. The 'exclude inode' option, if applied to a directory
941 inode, excludes now correctly all the leaves of this
942 directory. Thanks to John R. Dennison
943 <gerdesas@users.sourceforge.net> for reporting the bug.
945 4. Fixed the '-e' option to disable the possibility
946 to exclude the root inode (which causes the dumps to
947 be unreadable by restore). Prevented array overflow
948 when multiple -e options are used.
950 5. Fixed dump to correctly interpret a filesystem argument
951 which is a mountpoint and it is not an absolute path
952 (as specified in the fstab). Thanks to Bernhard R. Erdmann
953 <be@berdmann.de> for reporting the bug.
955 6. Made dump able to backup files larger than 2 GB. Note that
956 dump still doesn't cope with files larger than 4 GB.
958 7. Restore the real uid and gid before invoking an external
959 RSH program (big hole when dump or restore is suid root!).
961 8. Get the values for the default tape device and the location
962 of fstab file from the system headers. Thanks to
963 Andreas Dilger <adilger@turbolinux.com> for the patch.
965 9. Made dump -w|-W report all recognized filesystems
966 present in either /etc/fstab or /etc/dumpdates, and present
967 the list in the same order as in fstab file. Thanks
968 to Andreas Dilger <adilger@turbolinux.com> for the patch.
970 10. Made dump's -a (automatic end of tape detection) the
971 default. Specifying one of -B, -c, -d or -s options will
972 override the EOM detection. Thanks to Andreas Dilger
973 <adilger@turbolinux.com> for the patch.
975 11. Save the ext2 filesystem label into the dump volume label.
976 Specifying a volume label on the command line (-L option)
977 will override this feature. Thanks to Andreas Dilger
978 <adilger@turbolinux.com> for the patch.
980 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
981 =====================================================================
983 1. Fixed the signal handling in dump (which I broke in 0.4b17)
984 which was causing several strange dump failures (dump
985 hanged or segmentation faults etc).
987 2. Specified the default blocksize in the dump man page.
989 3. Changed two info messages of restore to be written on stdout
990 instead of stderr in order to leave stderr only for errors
991 or warnings. Thanks to Stephen Atwell
992 <satwell@urbana.css.mot.com> for the suggestion.
994 4. Corrected an off by one calculation which prevented
995 dumping a directory having a 1 character name length.
996 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
997 for reporting the bug.
999 5. Reinforce dump in order to allow only level 0 and no
1000 -u option when dumping a subdirectory, and document
1001 this in the man page. Thanks to Bernhard Erdmann
1002 <bernhard.erdmann@gmx.de> for reporting the bug.
1004 6. Fixed a small harmless bug in restore which caused
1005 in some conditions hard links to be restored several
1006 times (therefore generation some warning messages).
1007 Thanks to Kenneth Porter <shiva@well.com> for
1010 7. Updated the RPM spec file to the latest RedHat version,
1011 providing FHS packaging and other cosmetic changes.
1012 You will need RPM version >= 3.0.5 in order to build the RPM.
1014 8. Updated the configure script to check for libtermcap
1015 before checking for libreadline, because we need this
1016 library in order to compile the readline support.
1018 9. Made dump understand the LABEL= and UUID= notation
1019 both in /etc/fstab and on the command line. Note that
1020 you will need the /proc filesystem in order to use
1021 these notations. Thanks to Erik Troan <ewt@redhat.com>
1022 for providing the patch.
1024 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
1025 ===================================================================
1027 1. Fixed a potential buffer overflow in restore. Thanks
1028 to Stan Bubrouski <satan@fastdial.net> for reporting
1031 2. Fixed a readline-related bug which prevented
1032 'cat DUMPFILE | restore -i -f -' from working. Thanks
1033 to Charles Karney <karney@users.sourceforge.net>
1036 3. Changed a few "panic" into "exit", causing restore to
1037 be more stable against some attacks (like the last one
1038 reported on Bugtraq, although the last version of restore
1039 was not vulnerable - just dumped core). Thanks to
1040 Andreas Hasenack <andreas@conectiva.com.br> for reporting
1043 4. Removed the suid-root bit on dump and restore in the
1044 default build (and generated RPMs). It should be safer
1045 now. Users who need the suid-root capabilities in order
1046 to make network backups should read first the man page
1047 of dump and enable the suid bit manually.
1049 5. Added -ltermcap to the compile parameters for restore
1050 when using readline support, in order to make the compile
1051 process work on some systems (SuSE ?). Thanks to
1052 Patrik Schindler <poc@pocnet.net> for reporting the bug.
1054 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
1055 ====================================================================
1057 1. The -F script is called now *only* at the end of a tape,
1058 not at the end of the dump. Thanks to Christian Haul
1059 <haul@informatik.tu-darmstadt.de> for the bug report.
1061 Normally, the device name and the current volume number
1062 are passed on the command line of the script. However,
1063 if you want the old style script (with no arguments
1064 passed on the command line), you can enable it in
1065 configure with the --enable-oldstylefscript.
1067 2. Use posix signal handling to preserve dump functionality
1068 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
1071 3. Made the exit code of restore in case of a 'C'ompare
1072 command reflect the result of the compare. An exit status
1073 of 0 means the dump archive is correct, 1 means tape errors,
1074 2 means that some files were modified. Thanks to Kenneth Porter
1075 <shiva@well.com> for the suggestion.
1077 4. Made (finally) quotes work in interactive restore.
1079 5. Small fixes in order to allow dump to compile with a
1080 really old e2fsprogs version. Thanks to Ian Zimmerman
1081 <itz@speakeasy.org> for the bug report.
1083 6. Add GNU readline capabilities to interactive restore.
1084 Use configure's --enable-readline flag to enable this feature.
1085 Thanks to Patrik Schindler <poc@pocnet.net> for the
1088 7. Do the compare on the fly in restore 'C' mode (this will
1089 allow not to exhaust the available /tmp space when
1090 ccmparing large files). Thanks to Kenneth Porter
1091 <shiva@well.com> for the suggestion.
1093 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
1094 ====================================================================
1096 1. Fixed some several duplicate 'DUMP: DUMP:' in the
1099 2. Corrected the estimation of blocks to dump. Note that
1100 this is still wrong for large directory inodes, due
1101 to the size of a BSD directory entry on the tape
1102 (estimation is based on the size of the ext2 directory,
1103 which is smaller because it doesn't contain the null
1104 character at the end).
1106 3. Made dump report the total number of MB written to
1107 tape at the end of run. Thanks to W. Reilly Cooley
1108 <wcooley@nakedape.cc> for the patch.
1110 4. Added the -X option to restore giving the possibility
1111 to read the names of the files to be extracted/listed
1112 from a text file (in addition of the command line).
1113 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
1116 5. Added the device name and the current volume number
1117 as arguments to the end of tape script (dump -F option).
1119 6. Made the multi-volume dump work again (after having
1120 broken it in 0.4b15).
1122 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
1123 ===================================================================
1125 1. Added a prompt command in interactive restore mode. Thanks
1126 to Andreas Dilger <adilger@home.com> for the patch.
1128 2. Fixed a buffer overflow problem in dump (caused by
1129 not checking the size of the filesystem parameter).
1130 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
1131 reporting this on Bugtraq (and to several dump users
1132 who forwarded me his mail).
1134 3. Added the '-F script' option to dump in order to
1135 launch a script at the end of each tape (to be used
1136 with a tape changer, or to notify the sysadmin by
1139 4. Fixed a bug in restore compare code caused by the changes
1142 5. Fixed the treatment of options using the old BSD syntax
1143 in both dump and restore.
1145 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
1146 =======================================================================
1148 1. Fixed a bug in dump which may have caused invalid deleted
1149 directories to be dumped out if they were deleted after the
1150 mapping pass. This could occure on active filesystem and lead
1151 to heap corruption (causing dump malfunction in many possible ways).
1152 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
1153 bug and submitting the patch.
1155 2. Fixed the handling of the filesystem parameter in dump. A
1156 '/mnt/disk' parameter caused the disk contents to be dumped,
1157 but a '/mnt/disk/' parameter caused the mountpoint directory
1158 to be dumped (generally an empty directory).
1160 3. Improved the output of dump in order to tell which directory
1161 it is currently dumping (when dumping a subtree).
1163 4. Added the '-e' exclude inode option to dump. Thanks to
1164 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
1166 5. Added a REPORTING-BUGS file in order to provide a guide
1167 on how to correctly report a bug in dump/restore.
1169 6. Fixed a restore problem when restoring a hard link to an inode
1170 having the immutable or append-only attribute set. Thanks to
1171 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
1173 7. Fixed a compatibility problem between dumps made on little
1174 endian machines (the format was correct) and big endian
1175 machines (the format was incorrect). This fix break the
1176 compatibility with the older dumps made on big endian
1177 machines (sparc, m86k, ppc etc). For the first time in
1178 linux dump's history (I believe), the dumps made by *BSD,
1179 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
1180 so interchangeable. Thanks to Rob Cermak
1181 <cermak@ahab.rutgers.edu> for submitting the bug and
1182 helping me test the fix.
1184 8. Fixed the way dump reports the remaining percent/time, if
1185 the number of blocks actually dumped exceeds the estimated
1186 number of blocks. Thanks to Jean-Paul van der Jagt
1187 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
1189 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
1190 ======================================================================
1192 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
1193 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
1195 2. Removed the requirement to build the RPM as root from the
1196 spec file. Thanks to Christian Weisgerber
1197 <naddy@mips.rhein-neckar.de> for submitting this.
1199 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
1200 causing dump to block when dumping really huge filesystems.
1201 Many thanks to Patrik Schindler <poc@pocnet.net> for
1202 helping me find this bug.
1204 4. Fixed the treatment for an interrupt signal when dump access
1205 the remote tape through RSH. Thanks to Christian Weisgerber
1206 <naddy@mips.rhein-neckar.de> for providing the patch.
1208 5. Fixed a bug which was causing dump/restore to display
1209 garbage characters instead of the remote host name.
1211 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
1212 =====================================================================
1214 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
1215 <kukuk@suse.de> for submitting the patch.
1217 2. Fix for the exit code when using the size estimate option of
1218 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
1219 submitting the patch.
1221 3. Handle EINTR in atomical reads/writes in dump, which was causing
1222 dump to fail on some systems. Thanks to Eric Jergensen
1223 <eric@dvns.com> for reporting the bug and submitting the patch.
1225 4. Handle more than 16 characters for the device names in dumpdates.
1226 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
1227 tracking down the problem and proposing the solution.
1229 5. Fixed a bug in dump which prevented the creation of the
1230 dumpdates file when doing a 0-level dump without already
1231 having a dumpdates file. Thanks to Patrik Schindler
1232 <poc@pocnet.net> for reporting the bug.
1234 6. Changed the way dump 'S' flag reports the size estimate
1235 from number of blocks into bytes (making it compatible
1236 with the Solaris version, and simplifying things for
1237 amanda users). Thanks to Jason L Tibbitts III
1238 <tibbs@math.uh.edu> for reporting the bug.
1240 7. Fixed a compatibility problem in linux/alpha dump tape format.
1241 Now the linux/alpha dump are (again) compatible with the
1242 other dump formats. But this breaks compatibility with
1243 older dumps made on alpha. Thanks to Mike Tibor
1244 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
1246 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
1247 ======================================================================
1249 1. Added a '--enable-kerberos' to configure.
1251 2. Added a 'S' option to dump which determines the amount of space
1252 that is needed to perform the dump without actually doing it, similar
1253 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
1254 <cermak@ahab.rutgers.edu>.
1256 3. Added a 'M' multi-volume option to dump and restore which enables
1257 dumping to multiple files (useful when dumping to an ext2
1258 partition to make several dump files in order to bypass the 2GB
1259 file size limitation). The 'f' argument is treated as a prefix and
1260 the output files will be named <prefix>001, <prefix>002 etc. With
1261 the 'M' flag, restore automatically selects the right file without
1262 asking to enter a new tape each time.
1264 4. Fixed a memory leak which was causing dump to grow very big
1265 (270MB when dumping a 10GB filesystem...). Thanks to Jason
1266 Fearon <jasonf@netrider.org.au> for reporting the bug.
1268 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
1269 ======================================================================
1271 1. Make configure test if the system glob routines support
1272 extended syntax (ALTDIRFUNC). If not, use the internal glob
1273 routines instead of system ones. Thanks to Bernhard Sadlowski
1274 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
1275 and helping me resolve this and other minor libc5 compiling
1278 2. Fix a problem when dumping a ext2fs with the 'filetype'
1279 feature enabled. Thanks to Patrick J. LoPresti
1280 <patl@cag.lcs.mit.edu> for reporting the bug and to
1281 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
1283 3. Made the nodump flag work on directories. A directory which
1284 has the nodump flag gets never dumped, regardless of its
1287 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
1288 which allows dump on an active ext3 filesystem. However, this
1289 is a "quick and dirty" patch which enables backup of an ext3
1290 filesystem through the ext2 compatibility (by ignoring the
1291 NEEDS_RECOVERY bit). The journal file is not recognized and
1292 it is dumped (it should not).
1294 5. Test the superblock compatibility flags when dumping, in order
1295 to be sure that we know how to deal with specific features.
1297 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
1298 ====================================================================
1300 1. Use lchown instead of chown, fixing a possible security problem
1301 when restoring symlinks (a malicious user could use this
1302 to deliberately corrupt the ownership of important system files).
1303 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
1304 this and providing the patch.
1306 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
1307 ====================================================================
1309 1. Put dump sources under CVS, added Id tags in all files so
1310 one can use 'ident' on binary files.
1312 2. Added the dump/restore version in the usage text so one can
1313 easily verify the version he is using.
1315 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
1316 a va_start/va_end problem on linux-ppc (always call va_start
1317 va_end in pairs each time we use a vararg function).
1319 4. Added again the DT_* constants because old libc does not
1320 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
1321 for submitting the bug report.
1323 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
1324 this should enable dumping big (huge) filesystems.
1326 6. Added the RSH environment variable in order to be able to
1327 use a rsh replacement like ssh when doing remote backups (and
1328 bypass the security limitations of rcmd). Now you can do remote
1329 backups without being root (or making dump setuid root).
1331 7. Modified again the way dumpdates works. For incremental dumps,
1332 we need to read dumpdates even if we are not using 'u' option.
1333 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
1336 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
1337 ===================================================================
1339 1. Removed the 'k' flag from the restore 'about' text if kerberos
1340 was not compiled in.
1342 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
1343 to them (fsetflags takes a char*, setflags an open fd!).
1345 3. (f)setflags is called only if the flags aren't empty. If the
1346 file is a special file, a warning is printed, because changing
1347 flags implies opening the device. Normally, a special file
1348 should not have any flag... (Debian bug #29775, patch provided
1349 by Abhijit Dasgupta <abhijit@ans.net>).
1351 4. Made possible to dump a file system not mentioned in /etc/fstab.
1352 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
1354 5. Changed the default behaviour to not create dumpdates
1355 unless 'u' option is specified. Removed the old "debian-patch"
1356 which provided the same thing. (Debian bug #38136, #33818).
1358 6. Removed all those dump*announce, since they were getting old...
1360 7. Added warning messages if dumpdates does not exist and
1361 when an inferior level dump does not exist (except for a level 0
1364 8. Debugged the glob calls in interactive mode: restore used a
1365 dirent struct which was different from the /usr/include/dirent.h
1366 one (this used to work, is it a glibc2 change?), so none of the
1367 compat glob (which used /usr/include/dirent.h) or the system glob
1368 worked. Restore use now the system dirent (and the system
1369 DT_* constants), which are compatible with BSD ones.
1371 9. Added a configure flag (--with-dumpdatespath) to specify
1372 the location of dumpdates file. By default, it is
1375 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1376 included the current date/version in man pages.
1378 11. Corrected the estimation of remaining time when
1379 the operator doesn't change the tapes quickly enough. This
1380 was an old bug, I thought I corrected it, and discovered
1381 that in fact it was corrected in two different places, so
1382 the results canceled each other...
1384 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1385 ===================================================================
1387 1. Integrated multiple patches from RedHat, Debian and SuSE:
1389 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1391 - several fixes in the man pages.
1392 - update the default tape device to /dev/st0.
1393 - many updates for Linux Alpha (byte ordering, size_t etc).
1395 - use environment variable for TMPDIR (instead of /tmp).
1396 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1397 - workaround egcs bug (RedHat bugs #4281 and #2989).
1398 - wire $(OPT) throughout Makefile's.
1400 2. Upgrade the dump revision to 1, making possible to dump filesystems
1401 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1403 3. Fix some compile warnings, prototype all functions.
1405 4. Use glibc err/glob instead of internal compatibility
1406 routines (only if available).
1408 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1409 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1411 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1412 ======================================================================
1414 1. Integrated the changes from FreeBSD-3.1-RELEASE
1415 (mostly bug fixes, buffer overruns, dump has now an "automatic
1416 tape length calculation" flag, dump/restore can use kerberos now
1417 (this is NOT tested), use environment variables for TAPE and
1420 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1422 3. Corrected a bug in restore when using 'C' option with multi-volumes
1423 tapes (files splited accros two tapes give "size changed" errors
1426 4. Corrected the long standing bug when dumping multiple tapes.
1427 This works for me, needs further testing.
1429 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1430 ====================================================================
1432 1. Dump now runs correctly on kernels 2.1.x
1433 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1435 Changes between versions 0.4b2 and 0.4b3
1436 ========================================
1438 1. Use realpath() if available
1440 2. Report statistics
1442 Changes between versions 0.4b1 and 0.4b2
1443 ========================================
1445 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1448 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1450 3. Fixed configure.in to correctly handle --enable-rmt
1452 Changes between versions 0.3 and 0.4b1
1453 ======================================
1455 1. Integrated the changes from 4.4BSD-Lite2
1457 2. Integrated the patches from Debian and Red Hat
1459 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1461 4. Changed dump to use the Ext2fs library to get block addresses. This
1462 should solve the endianness problem on SparcLinux.
1464 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1465 distribution's one) to use autoconf
1467 6. Fixed a few minor bugs
1469 Changes between versions 0.2e and 0.2f
1470 ======================================
1472 1. Added the creation of named pipes (fifos) in restore.
1474 2. Added the -N flag in the restore manual page.
1476 3. Added the file kernel.patch which contains the llseek() optimization
1477 patch for 1.2.x kernels.
1479 4. Fixed a bug in the restoration of symbolic links: owner and group were
1482 5. Integrated some changes from FreeBSD 2.2.
1484 6. Added a call to ftruncate() after restoring each file to restore
1485 correctly files ending by a hole.
1487 Changes between versions 0.2d and 0.2e
1488 ======================================
1490 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1491 with owner = root (0) and group = root/wheel/whatever (0).
1493 Changes between versions 0.2c and 0.2d
1494 ======================================
1496 1. Dump is now able to backup 2GB+ filesystems.
1498 2. Dump and restore can now be linked as static binaries.
1500 Changes between versions 0.2b and 0.2c
1501 ======================================
1503 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1505 Changes between versions 0.2a and 0.2b
1506 ======================================
1508 1. Really fixed the bug that I should have corrected in 0.2a.
1510 2. Enabled optimization again.
1512 Changes between versions 0.2 and 0.2a
1513 =====================================
1515 1. Disabled the optimization during compilation.
1517 Changes between versions 0.1 and 0.2
1518 ====================================
1520 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1521 the fs_type field (actually, I modified the file fstab.c to make it
1522 use the mntent functions).
1524 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1525 made by dump should be readable by the BSD restore and backups made
1526 by the BSD dump should be readable by restore. Unfortunately, this
1527 means that the dump format has changed between version 0.1 and version
1530 3. Dump is now able to backup a subtree, it is no longer limited to whole
1531 filesystems like the BSD version.
1533 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1536 Changes between versions 0.0 and 0.1
1537 ====================================
1539 1. Now create links rdump and rrestore during the `make install' step.
1541 2. Linux port specific bugs added to the manual pages
1543 3. Incorrect estimation of the number of tapes blocks fixed when doing
1544 incremental backups.
1546 4. Better ls-like format in restore in interactive mode.
projects / dump.git / blob