1 $Id: CHANGES,v 1.283 2006/03/13 10:33:44 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 Changes between versions 0.4b40 and 0.4b41 (released January 2, 2006)
11 =====================================================================
13 1. Fix restore of dumped Access Control Lists. The previous
14 code for EA works fine, but ACLs needed conversion from the
15 ext2/3 disk format to posix_acl format before restoring.
17 2. Fix some issues with restoration of EA on big endian
20 3. Fix restore when the symtab is over 2GB in size.
22 4. Made the directory hash indexing an optional feature,
23 accessible by the '-H' option of restore, and disable it
26 5. Fixed dump to not include extended attributes information
27 in the toc (archive) file which confused restore -t.
29 Changes between versions 0.4b39 and 0.4b40 (released May 2, 2005)
30 =================================================================
32 1. Changed restore to emit warnings (instead of emitting a fatal
33 error) if a file (or a directory) is unavailable for a
34 comparision (if the user doesn't have the necessary permissions
35 to access it for example). Thanks to Kenneth Porter
36 <shiva@sewingwitch.com> for the bug report.
38 2. Re-done the 'do not save directory entries to non-dumped inodes
39 (excluded from dump)' feature. The previous implementation
40 worked well for excluded directories but not for regular files.
41 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
44 3. Fixed a bug in dump where the tape size was miscalculated when
45 the user used -d/-s to specify the tape characteristics. Thanks
46 to Philip Goisman <goisman@physics.arizona.edu> for reporting
49 4. Fixed another bug introduced in restore with the hashtree
50 implementation. This one caused restore to stop saying
51 "removenode: non-empty directory" in some cases.
53 5. Added support for dumping and restoring ext2/3 extended
54 attributes (EA), like the access control lists (ACL) or
55 the security labels used by SELinux.
57 Changes between versions 0.4b38 and 0.4b39 (released January 21, 2005)
58 ======================================================================
60 1. The newly added dump_on_cd_3 example was buggy, replace it
61 with an updated version from Andrew Basterfield
62 <bob@cemetery.homeunix.org>.
64 2. Made restore to chdir() back into the initial directory when
65 dumping core while aborting a comparision operation. The
66 previous behaviour was to write the corefile at the root of
67 the directory being compared, which could very well be
68 read only and preventing the corefile generation. Thanks
69 to Kenneth Porter <shiva@sewingwitch.com> for the bug report.
71 3. Silenced the failure to call fgetflags() when comparing an
72 entry which has no ext2 attributes (as in lsattr()).
74 4. Fix a brown paper bug in restore -C which broke restore and
75 caused modifications on the filesystem being compared
76 (directories containing a file with the same name as the
77 directory get renamed to RSTTMP...). Thanks to Kenneth Porter
78 <shiva@sewingwitch.com> for finding the bug and helping me
81 5. Made restore -C force the -N flag (no writing allowed on
82 the disk) in order to prevent more bugs like the above one.
84 Changes between versions 0.4b37 and 0.4b38 (released January 7, 2005)
85 =====================================================================
87 1. Fix a couple of troff syntax bugs in the man pages.
88 Thanks to Eric Raymond <esr@thyrsus.com> for the patch.
90 2. Made restore use either libncurses or libtermcap, depending
91 on which one is available at configure time.
93 3. Fixed restore negative size display bug when comparing a
94 dump containing files over 2GB. Thanks to Steve Bonds
95 <sbonds@users.sourceforge.net> for the bug report.
97 4. Do not save directory entries to non-dumped inodes
98 (excluded from dump). This will eliminate the 'missing
99 file' warnings when doing 'restore -C'.
101 5. Fix dump crash when backuping a huge (2TB) filesystem,
102 due to a bogus calculation on the inode map size.
103 Thanks to Kevin B. Haines <K.B.Haines@rl.ac.uk> for
104 submitting the bug and testing the fix.
106 6. Fix a problem in restore where the final \0 in the symbolic
107 link names could have been lost, generating corrupt filenames.
108 Thanks to Kyle Wilson <kyle.wilson@amd.com> for reporting the
111 7. Implemented a hash list for the directory names in restore.
112 The linear list used before caused problems in interactive
113 restores when dealing with directories having thousands of
114 entries. Thanks to Brian Ristuccia <bristuccia@starentnetworks.com>
115 for reporting the bug.
117 8. Improved restore -C, this time including the directory
118 attributes into the comparision.
120 9. Made restore understand tapes containing EA/ACLs (which will
121 be dumped by the next version of dump). In this version
122 extended attributes on the tape are ignored, for full EA/ACL
123 support wait for the next version or try the experimental EA
126 Changes between versions 0.4b36 and 0.4b37 (released July 7, 2004)
127 ==================================================================
129 1. Added the --enable-staticz configure option which enables
130 dump and restore to be linked with static versions of
131 libz and libbz2 (and dynamic versions of all other libraries).
132 This will make Debian users happy, because libz and libbz2
133 were the only needed libraries living in /usr, all the
134 others live in /lib. In case of system emergency, it is
135 better not to have to rely on an extra filesystem.
136 Thanks to Bdale Garbee <bdale@gag.com> for the suggestion.
138 2. Fix compilation on (at least the Linux Debian port to) AMD64.
139 (<ext2fs/ext2_types.h> defines some types (__s64 and __u64)
140 that are also defined by <linux/types.h> (<asm/types.h>) and
143 3. Make dump's reading of the dumpdates file a bit more robust,
144 preventing dump from crashing when the dumpdates file has
145 been modified by hand.
147 4. Fixed some offset calculations in dump code which could
148 lead to "bread lseek errors" on large filesystems. Thanks
149 to Bruce Lowekamp <lowekamp@users.sourceforge.net> for
150 reporting this bug and debugging the issue.
152 5. Made dump use the blkid library when searching for devices
153 by label or uuid instead of dump's own routines.
155 6. Corrected a bug in dump where a wrong LABEL=... line in
156 /etc/fstab could prevent dump from dumping unrelated
157 filesystems. Thanks to Bruce Lowekamp
158 <lowekamp@users.sourceforge.net> for reporting the bug.
160 Changes between versions 0.4b35 and 0.4b36 (released April 21, 2004)
161 ====================================================================
163 1. Fixed dump compilation with old gcc versions. Thanks to
164 Mike Castle <dalgoda@users.sourceforge.net> for the patch.
166 2. Fixed some warnings (howmany, roundup, powerof2 redefined)
167 when compiling against a recent glibc version.
169 3. Fixed a bug in restore preventing the read of a dump tape
170 written with Solaris 7 ufsdump. Thanks to Patrick Higgins
171 <phiggins@transzap.com> for reporting the bug and providing
174 4. Changed dump to enable the creation of volumes of different
175 sizes in a single run (make -B accept a list of values).
176 Patch contributed by Florian Zumbiehl <florz@gmx.de>.
178 5. Use the glibc provided minor() and major() macros instead
179 of our own bitmask implementation. This should be safe for
180 when the major/minor namespace will migrate to 32 bits.
181 Thanks to Zhang Jun <zhangjun@nanjing-fnst.com> for reporting
184 6. Made explicit in the dump man page that dump will not create
185 a remote file, it will only write to an already existing one.
187 7. Another try at making size estimates better again.
189 8. Put back the inconditional running of the end-of-tape
190 script which was changed in 0.434 to be run only when -M
191 or multiple -f were NOT specified. Some users rely on this
192 feature even when it is combined with -M/-f.
194 9. Fixed restore when restoring huge backups (where rstdir...
195 temporary files are over 2GB). Thanks to
196 Raphael Jaffey <rjaffey@artic.edu> for reporting this,
197 debugging the issue and making the patch.
199 10. Made restore understand FreeBSD UFS2 tapes. Thanks to
200 David <vrtin@users.sourceforge.net> for submitting the bug
201 and providing a test case.
203 11. Made dump work with an arbitrary integer level (no more 10
204 levels only limitation). Thanks to Matthew
205 <msvincen@midway.uchicago.edu> for the patch.
207 Changes between versions 0.4b34 and 0.4b35 (released December 21, 2003)
208 =======================================================================
210 1. Added a note in the dump man page saying that the default
211 blocksize can be 32 if -d is used with a high density argument.
212 Thanks to Antonios Christofides <A.Christofides@itia.ntua.gr>
215 2. Fixed configure to correctly understand CPPFLAGS, CFLAGS,
216 LDFLAGS environment variables. Thanks to Arcady Genkin
217 <antipode@users.sourceforge.net> for reporting the bug.
219 3. Made -e/-E options of dump accept an unlimited number of inodes
220 to be excluded, rather than a hardcoded maximum. Thanks to
221 Dietrich Rothe <d-rothe@users.sourceforge.net> for the patch.
223 4. Updated the autoconf system to 2.50. Forced the -D_BSD_SOURCE
224 and -D_USE_BSD_SIGNAL defines in configure in order to solve
225 64bit build problems because quad_t is redifined with a
226 different signature. Thanks to Mike Harris <mharris@redhat.com>
227 for reporting this bug.
229 5. Made restore build on Solaris, making possible to
230 restore Linux's "enhanced" tapes. Thanks to Uwe Gohlke
231 <uwe@ugsoft.de> for the patch.
233 6. Made an extension in the dump tape format capable of saving
234 MacOSX specific inode extensions. Uwe Gohlke <uwe@ugsoft.de>
235 wrote the extension and contributed the restore code back
236 into this codebase. The same extension mechanism will be
237 used in the future to save ACLs...
239 7. Made rmt work correctly with regard to QFA and local/remote
240 files and tapes. The remote access will however work only
241 when the dump provided rmt version is used. If you want to
242 use another rmt server, please do not use the QFA feature.
243 Thanks to Uwe Gohlke <uwe@ugsoft.de> for the patch.
245 Changes between versions 0.4b33 and 0.4b34 (released April 18, 2003)
246 ====================================================================
248 1. Fixed the INSTALL file to reflect the actual install paths.
249 Thanks to David Ronis <ronis@ronispc.chem.mcgill.ca> for
252 2. Fixed the configure script to only check for headers presence
253 instead of trying to compile them. This should fix issues
254 with old build environments. Thanks to Kari Lempiainen
255 <kari@funky.pp.fi> for reporting the bug.
257 3. Fixed restore to correctly ignore sockets when comparing
258 a dump (as socket cannot be properly restored anyway). Thanks
259 to Gunther Reiszig <gunther@mit.edu> for reporting the bug.
261 4. Fixed restore to correctly access the archive file (-A argument)
262 even when using a remote tape. Thanks to Fabrice Bellet
263 <fabrice@bellet.info> for reporting the bug.
265 5. Fixed (again) handling of long (largefile) seeks in rmt.
266 Thanks to Fabrice Bellet <fabrice@bellet.info> for reporting
269 6. Fixed restore corner case when dealing with large block sizes
270 dump is able to write now (-b 1024). Thanks to Fabrice Bellet
271 <fabrice@bellet.info> for reporting the bug.
273 7. Fixed a bug preventing dump to access a filesystem having
274 a label exactly 16 bytes in length. Thanks to <minduni@ti-edu.ch>
275 for reporting the bug.
277 8. Made dump store dump dates using explicit timezones, fixing a
278 problem with incremental dumps if the timezone is modified
279 between the dumps. Thanks to Philippe Troin <phil@fifi.org> for
280 the bug report and the patch.
282 9. Fixed a bug encountered when dumping individual files (not full
283 filesystems or directories) and dangling symbolic links happen
284 to be in the list of files. For as far as dump is concerned,
285 dangling symbolic links are allowed, and are dumped as is.
286 Thanks to Jin-su Ahn <jsahn@ee.snu.ac.kr> for reporting the
287 bug and providing the fix.
289 10. Fixed open and creation modes and permissions for QFA and
290 table-of-contents files in dump and restore. Thanks to
291 Philippe Troin <phil@fifi.org> for the patch.
293 11. Fixed the archive file descriptor handling enabling it to be 0.
294 This can happen in some cases when shell redirections are used.
295 Thanks to Philippe Troin <phil@fifi.org> for the patch.
297 12. Delayed the opening of archive file until after suid had been
298 dropped (fixing a possible security issue if dump is suid).
299 Thanks to Philippe Troin <phil@fifi.org> for the patch.
301 13. Fixed the 'S' command handling in the rmt client part.
302 Thanks to Philippe Troin <phil@fifi.org> for the patch.
304 14. Modified the end-of-tape script handling to print out statistics
305 (and stop the timer) before launching the eot script. Also, the eot
306 script does not get run anymore when using -M (which makes sense) or
307 when multiple tapes are listed on the command line
308 (-f tape0,tape1,tapen) (which also makes sense).
309 Thanks to Philippe Troin <phil@fifi.org> for the patch.
311 15. Relicensed dump/restore under the 'revised' BSD license, as per
312 ftp://ftp.cs.berkeley.edu/ucb/4bsd/README.Impt.License.Change.
314 16. Added LZO compression to dump. This new compression method has
315 the advantage of being super fast, thus not killing tape streaming
316 on slow machines. Thanks to Helmut Jarausch
317 <jarausch@igpm.rwth-aachen.de> for the patch and to
318 Markus Oberhumer <markus@oberhumer.com> for giving special permission
319 to include his miniLZO project (GPL licensed) in dump/restore.
321 17. Some small buffer overruns fixes in rmt. Thanks to Antonomasia
322 <ant@notatla.demon.co.uk> for reporting the bugs.
324 18. Added a special rmt version which can do encryption when writing
325 to tape. Read examples/encrypted_rmt/README for details on
326 how to enable and configure it. Thanks to Ken Lalonde
327 <ken@globalremit.com> for the patch.
329 19. Made dump work with 2.5 kernel end of tape early warning semantics.
330 Thanks to Kai Makisara <Kai.Makisara@kolumbus.fi> for the patch.
332 20. Fixed a bug which caused dump -w|-W not to work anymore, because
333 the fs_freq and fs_passno fields in /etc/mtab are always set
334 to 0 0. Thanks to Trent Piepho <xyzzy@speakeasy.org> for
337 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
338 =======================================================================
340 1. Added a note in the restore man page clarifying the question
341 'set the permissions on the current directory ?' asked by
342 restore at the end of treatment in -i and -x modes.
344 2. Fixed the endianess issues when reading compressed tapes.
345 Thanks to Dark Force <daq4th@users.sourceforge.net> for
346 reporting this bug and providing test cases.
348 3. Fixed the "ACL won't be dumped" warning message (which showed
349 an extra, unrelated error message). Thanks to Dragan Krnic
350 <dkrnic@lycos.com> for reporting this bug.
352 4. Made dump look first into /etc/mtab, then into /etc/fstab
353 when searching for filesystem paths. Also fixed some problems
354 caused by binding mounts or multiple block device mounts.
355 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
356 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
357 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
358 and #160305, Sourceforge bugs #618699 and #679832).
360 5. Made dump's -I option accept the value '0' meaning all the
361 read errors will be ignored. This can be useful when running
362 dump from unattended sessions (like cron jobs). Thanks to
363 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
365 6. Fixed the output of dump to indicate 'blocks' instead of
366 'tape blocks' in the various messages (blocks are always
367 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
368 made it clearly print the current blocksize at the start of
369 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
372 7. Made rmt understand long (largefiles) seeks.
374 8. Fixed build with very old versions of libext2fs, where
375 EXT2_FT_* constants were undefined.
377 9. Made dump accept the dumpdates path on the command line
378 (-D file option) instead of using only the hardcoded one.
379 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
382 10. Enabled rmt, LFS, readline, QFA options by default in
383 ./configure. Updated the configure process (new versions
384 of config.guess, config.sub etc).
386 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
387 =======================================================================
389 1. Changed dump to use fcntl(F_SETLK) style locking instead
390 of flock() when locking the dumpdates file. With the old
391 locking scheme, a local user having read rights on the
392 dumpdates file could be able to do a Denial of Service attack
393 on dump. In order to lock the dumpdates file with the new
394 scheme, the user would need to have write access on the file.
395 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
396 reporting the bug (originally a bugtraq post).
398 2. Fixed interactive 'ls' which caused spurious errors warnings
399 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
400 <jorgen_ostling@users.sourceforge.net> for reporting this
403 3. Fixed dump's estimate when dealing with sparse inodes.
405 4. Modified dump to allow setting a blocksize bigger than 32kB
406 (raised the limit to 1024kB), because newer hardware needs
407 this for proper operation. Thanks to Dirk Traenapp
408 <dtraenapp@users.sourceforge.net> for reporting this.
410 5. Fixed a bug causing Dump to stop and report an error if an
411 inode number in the exclude file was followed by some amount
412 of whitespace. Thanks to Jeffrey Sofferin
413 <sofferin@users.sourceforge.net> for reporting this bug.
415 6. Fixed a bug which caused restore, in some particular cases,
416 to ask some 'scary' questions and leave a bunch of RSTTMP
417 directories behind when restoring incremental tapes. Thanks
418 to Philippe Troin <phil@fifi.org> for reporting this bug and
419 providing the test cases.
421 7. Changed the wording when inodes are excluded from dump:
422 replaced 'Added inode 7 to exclude list' with
423 'Excluding inode 7 (resize inode) from dump', as suggested
424 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
426 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
427 ===================================================================
429 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
430 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
431 <eros@lamel.bo.cnr.it> for reporting the bug.
433 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
434 ===================================================================
436 1. Made dump print out the ext2 logical block number in case of a read
437 error, which can be given as an argument to debugfs' ncheck command,
438 in order to find out the affected inode. Added note about this
439 usage in dump's man page.
441 2. Fixed a problem in restore when reading tapes written on big
442 endian machines with very old versions of dump. The patch was
443 contributed by George Helffrich <george@geology.bristol.ac.uk>.
445 3. Fixed the tape length calculation when using large tapes
446 and compression. Thanks to Georg Lippold
447 <g_lippold@sourceforge.net> for reporting the bug.
449 4. Added a new set of examples in dump_on_cd_2 directory, based
450 on dump_on_cd examples but somewhat enhanced, supporting DVD
451 media, and localized in english and german. Thanks to
452 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
454 5. Made dump save 32 bit UID/GID. Internally, this was achieved
455 by switching from the old BSD inode format to the new BSD
456 inode format, which means that the tape format was changed.
457 However, since all restore versions out there should
458 transparently support both inode formats, the change should
459 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
462 6. Fixed a lot of warnings in the code shown when compiling
463 with 'gcc -W'. Thanks to Matthias Andree
464 <matthias.andree@stud.uni-dortmund.de> for reporting this.
466 7. Fixed a small markup bug in the dump man page. Thanks to
467 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
470 8. Rewrote entirely the man pages using the tmac.an macro
471 package (Linux man page format) instead of the original BSD
472 format. They should be now cleaner and easier to modify.
474 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
475 ==================================================================
477 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
478 non Linux clients were misinterpreted. The description of the problem
479 (incompatible numbering in Linux mtio opcodes) is documented at
480 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
481 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
482 providing an excellent, cross-platform replacement for rmt in his
485 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
486 which caused restore to fail when the inode maps were not entirely
487 contained on the first volume (when using really small volumes or
488 when dumping a huge number of inodes).
490 3. Fixed a problem in dump, where files in subdirectories of directories
491 that have the nodump flag set (or being excluded with -e/-E) were
492 sometimes incorrectly dumped (instead of being excluded). The original
493 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
494 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
495 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
496 help testing the patch.
498 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
499 which makes one able to pipe the output of dump, by the net, to
500 a remote CD-burner server.
502 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
503 and use GNU's symbolic syntax over rmt instead of numerical values
504 to assure multiple platform compatibility.
506 6. Documented the -d option in restore.
508 7. Added a -v (verbose) mode to dump. For now it just prints the number
509 of the inode being dumped, but this could evolve in future versions
510 to include interesting debugging output.
512 8. Added a -o flag to restore, which automatically restores the current
513 directory permissions in -i or -x mode, without asking the operator
514 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
515 and Peter Samuel <peters@e-smith.com>.
517 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
518 ====================================================================
520 1. Fixed a bug in the QFA feature which could have caused
521 a bad tape positionning on restore (causing extra delays in
522 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
524 2. Added a small note in the dump man page specifying that
525 there should be no white space between the option letter and
526 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
528 3. Made restore work with older versions of the readline library,
529 by checking in configure for several extended symbols. Restore
530 can now be compiled with a readline as old as the 2.0 release
531 (though it may be a good idea to upgrade it to a more recent
532 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
533 reporting the build failures.
535 4. Fixed a performance problem with the QFA file creation in
536 dump, which made unnecessary seeks on the tape slowing down
537 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
538 reporting this issue.
540 5. Removed the inclusion of some kernel headers in the dump
541 source, which prevented the compile in some kernel/glibc
542 headers/architecture combination. Thanks to Bdale Garbee
543 <bdale@gag.com> for reporting the bug.
545 6. Added the appropriate error message when dump fails to
546 open the output file for writing. Thanks to Amith Varghese
547 <amithv@yahoo.com> for reporting this bug.
549 7. Made restore able to understand large Solaris ufsdump tapes
550 (containing inodes bigger than 4194304). Sun have introduced
551 an "extension" to the dump tape format when dealing with
552 those inodes, which was uncorrectly handled by Linux restore.
553 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
554 providing a test case.
556 8. Added the -m parameter to dump which optimises the output for
557 inodes having been changed but not modified since the last dump
558 ('changed' and 'modified' have the meaning defined in stat(2)).
559 For those inodes, dump will save only the metadata, instead of
560 saving the entire inode contents. Inodes which are either
561 directories or have been modified since the last dump are saved
562 in a regular way. Uses of this flag must be consistent, meaning
563 that either every dump in an incremental dump set have the flag,
566 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
567 =======================================================================
569 1. Fixed behaviour of dump when exceeding resource limits
572 2. Added the -L flag to restore to allow the user to specify a
573 maximal allowed number of miscompares when using restore
574 with the -C option to check the backup.
576 3. Detailed the manual entry for the -N option of restore.
578 4. Added the -a flag to restore to make able doing unattended
579 restores in -i or -x mode (automatically walks through the
580 multiple dump volumes).
582 5. Extended the QFA mode to work with local files and/or
583 remote tapes and files. This way, restore can know in advance
584 the tape number and the offset for the inodes to extract and
585 can minimize the extraction time by seeking directly to the
588 6. Added the -A <archive> option to both dump and restore,
589 which makes dump to archive a dump table-of-contents in
590 the specified file to be used by restore to determine
591 whether a file is in the dump file that is being restored.
592 (the archive file syntax is also compatible with the
593 Solaris ufsdump generated one).
595 7. Small fix in restore making it able to read some (broken ?)
596 Solaris ufsdump tapes.
598 8. Fixed dump to correctly recognise the root filesystem when using
599 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
600 for reporting this bug.
602 9. Added the -P <file> option to restore to create a
603 Quick File Access file from an already made dump. Patch
604 contributed by Uwe Gohlke <uwe@ugsoft.de>.
606 10. Made restore compile and run on Solaris, making it a
607 possible replacement for the standard ufsrestore. Port was
608 contributed by Uwe Gohlke <uwe@ugsoft.de>.
610 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
611 =====================================================================
613 1. Added a set of backup scripts from Eugenio Diaz
614 <getnito@yahoo.com> in the examples section. It features
615 automatic (cron based) full and incremental dumping of
616 several filesystems on a separate filesystem.
618 2. Fixed a off-by-one miscalculation which disabled dumping
619 a one letter subdirectory. Thanks to Chris Siebenmann
620 <cks@utcc.utoronto.ca> for reporting the bug.
622 3. Fixed several restore bugs occuring when trying to
623 restore some missing files on the tape. Thanks to Chris
624 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
626 4. Fixed --with-ldopts configure argument passing, installing from
627 a separate object directory, makefile cleanups contributed
628 by <splite@purdue.edu>.
630 5. Fix a bug which could caused, in some conditions, the highest
631 number inode of a filesystem, to not be dumped. Many thanks
632 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
635 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
636 =======================================================================
638 1. Added a mini howto from Patrick Walsh in the examples
641 2. Minor man pages syntax corrections. Thanks to
642 Chris Danis <screechco@home.com> for reporting the bugs.
644 3. Added a script from David B. Peterson <dave@toppledwagon.com>
645 to the examples section. It features dumping several
646 filesystems to a remote tape drive upon ssh.
648 4. Added a patch provided by Richard Jones <rich@annexia.org>
649 which allows BRADEMAX (number of read errors tolerated by
650 dump) to be adjusted using the -I option.
652 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
653 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
656 6. Add the -l option to restore to specify if, when doing a
657 remote restore, the file used to access the backup is a
658 regular file (the defaults being a tape drive). Restore needs
659 to know this information when reading a remote compressed
660 dump. Previously, this information was autodetected, but
661 the autodetection code fails (with ioctl: Inappropriate ioctl
662 for device) when using a non Linux remote box. Thanks to
663 many users and especially to Eros Albertazzi
664 <eros@lamel.bo.cnr.it> for reporting this.
666 7. Found a workaroung for the dump deadlock problem (3 childs
667 stuck in pause(), father in read()). The workaround seems
668 to work for me and several beta-testers. If it doesn't work
669 for you, please report back.
671 8. Updated the RPM spec file (BuildPrereq, URL etc).
673 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
674 ========================================================================
676 1. Fixed the permissions of a newly created QFA file by dump.
678 2. Cleaned up the source of dump (the external variables
679 definition was a complete mess, making possible to have
682 3. Fixed restore to use the full tape volume path when doing
683 a compare (since it changes the working directory to the
684 filesystem being compared in the process).
686 4. Added the -q option to dump which makes dump abort
687 whenever it needs operator attention. It should be
688 easier to use dump in scripts with this option.
690 5. Detect the use of incompatible options to dump and
691 refuse them (like -a and -B options together).
693 6. Added bzip2 compression to dump/restore (use option -j level
694 to select it). Note that you will need a restore version
695 >= 0.4b24 in order to restore a bzip2 compressed dump.
696 The same warning as for the zlib compression applies:
697 the tape format of a bzip2 dump is not compatible with the
698 original BSD tape format.
700 7. Fixed a overflow problem in dump corrupting the dump when
701 very large files were encountered. Thanks to Vladimir Nadvornik
704 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
705 kernel buffer/page cache before starting the dump, helping
706 a bit those who use dump on mounted filesystems. Thanks to
707 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
710 9. Updated the RPM spec file following the RedHat changes
711 (dynamically linked binaries now in /usr/sbin etc).
713 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
714 which enables restore to recognise multi volume compressed dumps
715 done on CD. Included his scripts for dump (which pipe the dump
716 output directly into cdrecord) and restore. There is now
717 possible to to backups to CD on the fly!
719 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
720 ===================================================================
722 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
723 Marcus Meissner (Caldera International Security Dept.).
725 2. Implement the Sun rmt extended protocol. Patch submitted
726 by Ian Gordon <iangordon@users.sourceforge.net>.
728 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
729 the linux kernel header. This ensures that dump always has the
730 latest version of this file. Patch submitted by
731 Andreas Dilger <adilger@turbolinux.com>.
733 4. Report any filesystem present in either /etc/fstab with a
734 positive passno or /etc/dumpdates in dump -w output.
735 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
737 5. Fixed the looping problem in dump introduced in the
740 6. Changed the -B option of dump to limit the size of
741 _compressed_ output per volume if compression is on.
742 Patch contributed by Helmut Jarausch
743 <jarausch@igpm.rwth-aachen.de>. Note however that, since
744 it is impossible to predict the size of the compressed
745 data before doing the compression, the -B limit is a bit
748 7. Fixed a bug in reading the operator typed file/tape path for
749 the next volume in restore.
751 8. Implemented a "-F script" option to restore which permits the
752 user to specify a script which will be launched at the
753 beginning of each tape, useful for automatic programming of
754 tape changers for example. See the restore man page for the
755 script parameters and return codes.
757 9. Small fix for the QFA routines provided by Uwe Gohlke
758 <uwe@ugsoft.de>, and some recommendations for QFA uses in
761 10. Fixed the multivolume restoring where making a mistake
762 to the 'Mount next tape' prompt caused several blocks to
765 11. Enhanced the -e option of dump to take as a parameter a
766 comma separated list of inode numbers.
768 12. Added the -E option to dump which specify a file containing
769 inode numbers to exclude from the dump.
771 13. Fixed the compressed multi-volume dump + restore.
773 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
774 ==================================================================
776 1. Made dump report the number of blocks dumped per volume.
777 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
779 2. Fix a bug in dump where waiting too much at the 'change volumes'
780 question caused the volume to be rewritten. Thanks to
781 Graham Reed <greed@users.sourceforge.net> for reporting the
782 bug and providing a patch.
784 3. Added a compression option to dump, contributed by Jerry
785 Peters <gapeters@worldnet.att.net>.
787 WARNING: the compression works only on files/pipes and
788 drives supporting variable block size.
790 WARNING2: tapes created using the compression feature are
791 incompatible with the standard dump tape format, and a
792 version of restore >= 0.4b22 will be needed for extraction.
794 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
795 architectures. Thanks to Paul Slootman <paul@debian.org> for
796 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
799 5. Many cleanups (CPP defines, const char warnings, check of
800 ext2fs COMPAT flags, time_t cleanups, added libext2 version
801 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
803 6. Made --prefix option work in configure. All the install path
804 are now based on the configure parameters.
806 7. Added the Quick File Access mode in dump/restore, contributed
807 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
808 a file tape position for each inode, and this file is used by
809 restore (if called with parameter Q and the filename)
810 to directly position the tape at the file restore is currently
811 working on. This saves hours when restoring single files from
812 large backups, saves the tapes and the drive's head. Use
813 --enable-qfa option of configure to compile in the QFA support.
815 8. Added the possibility to dump several files and directories
816 in a single invocation of dump. Thanks to Uwe Gohlke
817 <uwe@ugsoft.de> for implementing this option.
819 9. Fixed the dumping and restoring of files with holes
820 (files ending with holes didn't get dumped or restored
823 10. Fixed a socket descriptor leak in dump, which leaved opened
824 3 file descriptors per dump process (and there is one dump
827 11. Fixed dump large file system support, by explicit use of
828 open64/lseek64/etc functions (explicit use needed because
829 e2fsprogs libraries don't behave well when compiled with
830 FILE_OFFSET_BITS=64).
832 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
833 ======================================================================
835 1. Fixed some bugs in the dump -w|-W logic introduced by
836 the previous version. Thanks to Andreas Dilger
837 <adilger@turbolinux.com> for his help on this one.
839 2. Fixed again a compilation problem when using old e2fs
840 headers (filesystem label related). Thanks to many users
841 who reported this stupid error.
843 3. Fixed a build problem on old lib5 based systems dealing with
844 _PATH_MNTTAB being undefined. Thanks to John Adams
845 <johna@onevista.com> for reporting the bug.
847 4. Improved the error detection in restore when creating
848 the temporary files in TMPDIR. Now restore will corectly
849 report a 'no space left on device' error instead of
850 strange errors which could imply an archive corruption.
851 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
852 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
854 5. Added the throughput information in dump's progression
855 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
858 6. Use libext2fs's inode scanning routines, which are particularly
859 robust when dealing with errors (especially when having some
860 bad blocks on the filesystem). Many thanks to Colin
861 <colin@horizon.com> for reporting the bug, investigating
862 the issues, finding a workaround, writing the patch and
863 fully testing it... (of course, if this patch will break
864 anything, he is to blame for :-)).
866 7. Made dump and restore LFS aware. Dump can dump a filesystem
867 containing large files, generate a large file on output and
868 restore can restore them. This feature is not enabled by
869 default, use --enable-largefile option of configure to enable
870 it (you need to have a LFS aware glibc though). Thanks to
871 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
872 and to Theodore T'so <tytso@valinux.com> for his always
875 8. Made dump ask upon a tape write error if it should rewrite
876 the current volume (assume this is a bad tape) or if it should
877 assume an end-of-tape condition (useful for tape drives which
878 behaves badly at the end of the tape). Thanks to Andreas
879 Dilger <adilger@turbolinux.com> for the suggestion.
881 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
882 =======================================================================
884 1. Fixed a small compilation problem due to a change
885 in the definintion of the struct sigaction in
886 glibc 2.0 / libc5. Thanks to Gunther Schlegel
887 <schlegel@riege.de> for reporting the bug and to
888 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
891 2. Modified the label and uuid analysis in order to be
892 self-contained (independant of kernel/libc headers). This
893 should fix the compile with older kernel/libc headers and
894 will preserve the functionality. Thanks to Bernhard Erdmann
895 <bernhard.erdmann@gmx.de> for reporting the bug.
897 3. The 'exclude inode' option, if applied to a directory
898 inode, excludes now correctly all the leaves of this
899 directory. Thanks to John R. Dennison
900 <gerdesas@users.sourceforge.net> for reporting the bug.
902 4. Fixed the '-e' option to disable the possibility
903 to exclude the root inode (which causes the dumps to
904 be unreadable by restore). Prevented array overflow
905 when multiple -e options are used.
907 5. Fixed dump to correctly interpret a filesystem argument
908 which is a mountpoint and it is not an absolute path
909 (as specified in the fstab). Thanks to Bernhard R. Erdmann
910 <be@berdmann.de> for reporting the bug.
912 6. Made dump able to backup files larger than 2 GB. Note that
913 dump still doesn't cope with files larger than 4 GB.
915 7. Restore the real uid and gid before invoking an external
916 RSH program (big hole when dump or restore is suid root!).
918 8. Get the values for the default tape device and the location
919 of fstab file from the system headers. Thanks to
920 Andreas Dilger <adilger@turbolinux.com> for the patch.
922 9. Made dump -w|-W report all recognized filesystems
923 present in either /etc/fstab or /etc/dumpdates, and present
924 the list in the same order as in fstab file. Thanks
925 to Andreas Dilger <adilger@turbolinux.com> for the patch.
927 10. Made dump's -a (automatic end of tape detection) the
928 default. Specifying one of -B, -c, -d or -s options will
929 override the EOM detection. Thanks to Andreas Dilger
930 <adilger@turbolinux.com> for the patch.
932 11. Save the ext2 filesystem label into the dump volume label.
933 Specifying a volume label on the command line (-L option)
934 will override this feature. Thanks to Andreas Dilger
935 <adilger@turbolinux.com> for the patch.
937 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
938 =====================================================================
940 1. Fixed the signal handling in dump (which I broke in 0.4b17)
941 which was causing several strange dump failures (dump
942 hanged or segmentation faults etc).
944 2. Specified the default blocksize in the dump man page.
946 3. Changed two info messages of restore to be written on stdout
947 instead of stderr in order to leave stderr only for errors
948 or warnings. Thanks to Stephen Atwell
949 <satwell@urbana.css.mot.com> for the suggestion.
951 4. Corrected an off by one calculation which prevented
952 dumping a directory having a 1 character name length.
953 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
954 for reporting the bug.
956 5. Reinforce dump in order to allow only level 0 and no
957 -u option when dumping a subdirectory, and document
958 this in the man page. Thanks to Bernhard Erdmann
959 <bernhard.erdmann@gmx.de> for reporting the bug.
961 6. Fixed a small harmless bug in restore which caused
962 in some conditions hard links to be restored several
963 times (therefore generation some warning messages).
964 Thanks to Kenneth Porter <shiva@well.com> for
967 7. Updated the RPM spec file to the latest RedHat version,
968 providing FHS packaging and other cosmetic changes.
969 You will need RPM version >= 3.0.5 in order to build the RPM.
971 8. Updated the configure script to check for libtermcap
972 before checking for libreadline, because we need this
973 library in order to compile the readline support.
975 9. Made dump understand the LABEL= and UUID= notation
976 both in /etc/fstab and on the command line. Note that
977 you will need the /proc filesystem in order to use
978 these notations. Thanks to Erik Troan <ewt@redhat.com>
979 for providing the patch.
981 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
982 ===================================================================
984 1. Fixed a potential buffer overflow in restore. Thanks
985 to Stan Bubrouski <satan@fastdial.net> for reporting
988 2. Fixed a readline-related bug which prevented
989 'cat DUMPFILE | restore -i -f -' from working. Thanks
990 to Charles Karney <karney@users.sourceforge.net>
993 3. Changed a few "panic" into "exit", causing restore to
994 be more stable against some attacks (like the last one
995 reported on Bugtraq, although the last version of restore
996 was not vulnerable - just dumped core). Thanks to
997 Andreas Hasenack <andreas@conectiva.com.br> for reporting
1000 4. Removed the suid-root bit on dump and restore in the
1001 default build (and generated RPMs). It should be safer
1002 now. Users who need the suid-root capabilities in order
1003 to make network backups should read first the man page
1004 of dump and enable the suid bit manually.
1006 5. Added -ltermcap to the compile parameters for restore
1007 when using readline support, in order to make the compile
1008 process work on some systems (SuSE ?). Thanks to
1009 Patrik Schindler <poc@pocnet.net> for reporting the bug.
1011 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
1012 ====================================================================
1014 1. The -F script is called now *only* at the end of a tape,
1015 not at the end of the dump. Thanks to Christian Haul
1016 <haul@informatik.tu-darmstadt.de> for the bug report.
1018 Normally, the device name and the current volume number
1019 are passed on the command line of the script. However,
1020 if you want the old style script (with no arguments
1021 passed on the command line), you can enable it in
1022 configure with the --enable-oldstylefscript.
1024 2. Use posix signal handling to preserve dump functionality
1025 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
1028 3. Made the exit code of restore in case of a 'C'ompare
1029 command reflect the result of the compare. An exit status
1030 of 0 means the dump archive is correct, 1 means tape errors,
1031 2 means that some files were modified. Thanks to Kenneth Porter
1032 <shiva@well.com> for the suggestion.
1034 4. Made (finally) quotes work in interactive restore.
1036 5. Small fixes in order to allow dump to compile with a
1037 really old e2fsprogs version. Thanks to Ian Zimmerman
1038 <itz@speakeasy.org> for the bug report.
1040 6. Add GNU readline capabilities to interactive restore.
1041 Use configure's --enable-readline flag to enable this feature.
1042 Thanks to Patrik Schindler <poc@pocnet.net> for the
1045 7. Do the compare on the fly in restore 'C' mode (this will
1046 allow not to exhaust the available /tmp space when
1047 ccmparing large files). Thanks to Kenneth Porter
1048 <shiva@well.com> for the suggestion.
1050 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
1051 ====================================================================
1053 1. Fixed some several duplicate 'DUMP: DUMP:' in the
1056 2. Corrected the estimation of blocks to dump. Note that
1057 this is still wrong for large directory inodes, due
1058 to the size of a BSD directory entry on the tape
1059 (estimation is based on the size of the ext2 directory,
1060 which is smaller because it doesn't contain the null
1061 character at the end).
1063 3. Made dump report the total number of MB written to
1064 tape at the end of run. Thanks to W. Reilly Cooley
1065 <wcooley@nakedape.cc> for the patch.
1067 4. Added the -X option to restore giving the possibility
1068 to read the names of the files to be extracted/listed
1069 from a text file (in addition of the command line).
1070 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
1073 5. Added the device name and the current volume number
1074 as arguments to the end of tape script (dump -F option).
1076 6. Made the multi-volume dump work again (after having
1077 broken it in 0.4b15).
1079 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
1080 ===================================================================
1082 1. Added a prompt command in interactive restore mode. Thanks
1083 to Andreas Dilger <adilger@home.com> for the patch.
1085 2. Fixed a buffer overflow problem in dump (caused by
1086 not checking the size of the filesystem parameter).
1087 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
1088 reporting this on Bugtraq (and to several dump users
1089 who forwarded me his mail).
1091 3. Added the '-F script' option to dump in order to
1092 launch a script at the end of each tape (to be used
1093 with a tape changer, or to notify the sysadmin by
1096 4. Fixed a bug in restore compare code caused by the changes
1099 5. Fixed the treatment of options using the old BSD syntax
1100 in both dump and restore.
1102 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
1103 =======================================================================
1105 1. Fixed a bug in dump which may have caused invalid deleted
1106 directories to be dumped out if they were deleted after the
1107 mapping pass. This could occure on active filesystem and lead
1108 to heap corruption (causing dump malfunction in many possible ways).
1109 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
1110 bug and submitting the patch.
1112 2. Fixed the handling of the filesystem parameter in dump. A
1113 '/mnt/disk' parameter caused the disk contents to be dumped,
1114 but a '/mnt/disk/' parameter caused the mountpoint directory
1115 to be dumped (generally an empty directory).
1117 3. Improved the output of dump in order to tell which directory
1118 it is currently dumping (when dumping a subtree).
1120 4. Added the '-e' exclude inode option to dump. Thanks to
1121 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
1123 5. Added a REPORTING-BUGS file in order to provide a guide
1124 on how to correctly report a bug in dump/restore.
1126 6. Fixed a restore problem when restoring a hard link to an inode
1127 having the immutable or append-only attribute set. Thanks to
1128 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
1130 7. Fixed a compatibility problem between dumps made on little
1131 endian machines (the format was correct) and big endian
1132 machines (the format was incorrect). This fix break the
1133 compatibility with the older dumps made on big endian
1134 machines (sparc, m86k, ppc etc). For the first time in
1135 linux dump's history (I believe), the dumps made by *BSD,
1136 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
1137 so interchangeable. Thanks to Rob Cermak
1138 <cermak@ahab.rutgers.edu> for submitting the bug and
1139 helping me test the fix.
1141 8. Fixed the way dump reports the remaining percent/time, if
1142 the number of blocks actually dumped exceeds the estimated
1143 number of blocks. Thanks to Jean-Paul van der Jagt
1144 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
1146 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
1147 ======================================================================
1149 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
1150 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
1152 2. Removed the requirement to build the RPM as root from the
1153 spec file. Thanks to Christian Weisgerber
1154 <naddy@mips.rhein-neckar.de> for submitting this.
1156 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
1157 causing dump to block when dumping really huge filesystems.
1158 Many thanks to Patrik Schindler <poc@pocnet.net> for
1159 helping me find this bug.
1161 4. Fixed the treatment for an interrupt signal when dump access
1162 the remote tape through RSH. Thanks to Christian Weisgerber
1163 <naddy@mips.rhein-neckar.de> for providing the patch.
1165 5. Fixed a bug which was causing dump/restore to display
1166 garbage characters instead of the remote host name.
1168 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
1169 =====================================================================
1171 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
1172 <kukuk@suse.de> for submitting the patch.
1174 2. Fix for the exit code when using the size estimate option of
1175 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
1176 submitting the patch.
1178 3. Handle EINTR in atomical reads/writes in dump, which was causing
1179 dump to fail on some systems. Thanks to Eric Jergensen
1180 <eric@dvns.com> for reporting the bug and submitting the patch.
1182 4. Handle more than 16 characters for the device names in dumpdates.
1183 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
1184 tracking down the problem and proposing the solution.
1186 5. Fixed a bug in dump which prevented the creation of the
1187 dumpdates file when doing a 0-level dump without already
1188 having a dumpdates file. Thanks to Patrik Schindler
1189 <poc@pocnet.net> for reporting the bug.
1191 6. Changed the way dump 'S' flag reports the size estimate
1192 from number of blocks into bytes (making it compatible
1193 with the Solaris version, and simplifying things for
1194 amanda users). Thanks to Jason L Tibbitts III
1195 <tibbs@math.uh.edu> for reporting the bug.
1197 7. Fixed a compatibility problem in linux/alpha dump tape format.
1198 Now the linux/alpha dump are (again) compatible with the
1199 other dump formats. But this breaks compatibility with
1200 older dumps made on alpha. Thanks to Mike Tibor
1201 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
1203 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
1204 ======================================================================
1206 1. Added a '--enable-kerberos' to configure.
1208 2. Added a 'S' option to dump which determines the amount of space
1209 that is needed to perform the dump without actually doing it, similar
1210 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
1211 <cermak@ahab.rutgers.edu>.
1213 3. Added a 'M' multi-volume option to dump and restore which enables
1214 dumping to multiple files (useful when dumping to an ext2
1215 partition to make several dump files in order to bypass the 2GB
1216 file size limitation). The 'f' argument is treated as a prefix and
1217 the output files will be named <prefix>001, <prefix>002 etc. With
1218 the 'M' flag, restore automatically selects the right file without
1219 asking to enter a new tape each time.
1221 4. Fixed a memory leak which was causing dump to grow very big
1222 (270MB when dumping a 10GB filesystem...). Thanks to Jason
1223 Fearon <jasonf@netrider.org.au> for reporting the bug.
1225 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
1226 ======================================================================
1228 1. Make configure test if the system glob routines support
1229 extended syntax (ALTDIRFUNC). If not, use the internal glob
1230 routines instead of system ones. Thanks to Bernhard Sadlowski
1231 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
1232 and helping me resolve this and other minor libc5 compiling
1235 2. Fix a problem when dumping a ext2fs with the 'filetype'
1236 feature enabled. Thanks to Patrick J. LoPresti
1237 <patl@cag.lcs.mit.edu> for reporting the bug and to
1238 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
1240 3. Made the nodump flag work on directories. A directory which
1241 has the nodump flag gets never dumped, regardless of its
1244 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
1245 which allows dump on an active ext3 filesystem. However, this
1246 is a "quick and dirty" patch which enables backup of an ext3
1247 filesystem through the ext2 compatibility (by ignoring the
1248 NEEDS_RECOVERY bit). The journal file is not recognized and
1249 it is dumped (it should not).
1251 5. Test the superblock compatibility flags when dumping, in order
1252 to be sure that we know how to deal with specific features.
1254 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
1255 ====================================================================
1257 1. Use lchown instead of chown, fixing a possible security problem
1258 when restoring symlinks (a malicious user could use this
1259 to deliberately corrupt the ownership of important system files).
1260 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
1261 this and providing the patch.
1263 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
1264 ====================================================================
1266 1. Put dump sources under CVS, added Id tags in all files so
1267 one can use 'ident' on binary files.
1269 2. Added the dump/restore version in the usage text so one can
1270 easily verify the version he is using.
1272 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
1273 a va_start/va_end problem on linux-ppc (always call va_start
1274 va_end in pairs each time we use a vararg function).
1276 4. Added again the DT_* constants because old libc does not
1277 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
1278 for submitting the bug report.
1280 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
1281 this should enable dumping big (huge) filesystems.
1283 6. Added the RSH environment variable in order to be able to
1284 use a rsh replacement like ssh when doing remote backups (and
1285 bypass the security limitations of rcmd). Now you can do remote
1286 backups without being root (or making dump setuid root).
1288 7. Modified again the way dumpdates works. For incremental dumps,
1289 we need to read dumpdates even if we are not using 'u' option.
1290 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
1293 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
1294 ===================================================================
1296 1. Removed the 'k' flag from the restore 'about' text if kerberos
1297 was not compiled in.
1299 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
1300 to them (fsetflags takes a char*, setflags an open fd!).
1302 3. (f)setflags is called only if the flags aren't empty. If the
1303 file is a special file, a warning is printed, because changing
1304 flags implies opening the device. Normally, a special file
1305 should not have any flag... (Debian bug #29775, patch provided
1306 by Abhijit Dasgupta <abhijit@ans.net>).
1308 4. Made possible to dump a file system not mentioned in /etc/fstab.
1309 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
1311 5. Changed the default behaviour to not create dumpdates
1312 unless 'u' option is specified. Removed the old "debian-patch"
1313 which provided the same thing. (Debian bug #38136, #33818).
1315 6. Removed all those dump*announce, since they were getting old...
1317 7. Added warning messages if dumpdates does not exist and
1318 when an inferior level dump does not exist (except for a level 0
1321 8. Debugged the glob calls in interactive mode: restore used a
1322 dirent struct which was different from the /usr/include/dirent.h
1323 one (this used to work, is it a glibc2 change?), so none of the
1324 compat glob (which used /usr/include/dirent.h) or the system glob
1325 worked. Restore use now the system dirent (and the system
1326 DT_* constants), which are compatible with BSD ones.
1328 9. Added a configure flag (--with-dumpdatespath) to specify
1329 the location of dumpdates file. By default, it is
1332 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1333 included the current date/version in man pages.
1335 11. Corrected the estimation of remaining time when
1336 the operator doesn't change the tapes quickly enough. This
1337 was an old bug, I thought I corrected it, and discovered
1338 that in fact it was corrected in two different places, so
1339 the results canceled each other...
1341 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1342 ===================================================================
1344 1. Integrated multiple patches from RedHat, Debian and SuSE:
1346 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1348 - several fixes in the man pages.
1349 - update the default tape device to /dev/st0.
1350 - many updates for Linux Alpha (byte ordering, size_t etc).
1352 - use environment variable for TMPDIR (instead of /tmp).
1353 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1354 - workaround egcs bug (RedHat bugs #4281 and #2989).
1355 - wire $(OPT) throughout Makefile's.
1357 2. Upgrade the dump revision to 1, making possible to dump filesystems
1358 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1360 3. Fix some compile warnings, prototype all functions.
1362 4. Use glibc err/glob instead of internal compatibility
1363 routines (only if available).
1365 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1366 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1368 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1369 ======================================================================
1371 1. Integrated the changes from FreeBSD-3.1-RELEASE
1372 (mostly bug fixes, buffer overruns, dump has now an "automatic
1373 tape length calculation" flag, dump/restore can use kerberos now
1374 (this is NOT tested), use environment variables for TAPE and
1377 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1379 3. Corrected a bug in restore when using 'C' option with multi-volumes
1380 tapes (files splited accros two tapes give "size changed" errors
1383 4. Corrected the long standing bug when dumping multiple tapes.
1384 This works for me, needs further testing.
1386 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1387 ====================================================================
1389 1. Dump now runs correctly on kernels 2.1.x
1390 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1392 Changes between versions 0.4b2 and 0.4b3
1393 ========================================
1395 1. Use realpath() if available
1397 2. Report statistics
1399 Changes between versions 0.4b1 and 0.4b2
1400 ========================================
1402 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1405 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1407 3. Fixed configure.in to correctly handle --enable-rmt
1409 Changes between versions 0.3 and 0.4b1
1410 ======================================
1412 1. Integrated the changes from 4.4BSD-Lite2
1414 2. Integrated the patches from Debian and Red Hat
1416 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1418 4. Changed dump to use the Ext2fs library to get block addresses. This
1419 should solve the endianness problem on SparcLinux.
1421 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1422 distribution's one) to use autoconf
1424 6. Fixed a few minor bugs
1426 Changes between versions 0.2e and 0.2f
1427 ======================================
1429 1. Added the creation of named pipes (fifos) in restore.
1431 2. Added the -N flag in the restore manual page.
1433 3. Added the file kernel.patch which contains the llseek() optimization
1434 patch for 1.2.x kernels.
1436 4. Fixed a bug in the restoration of symbolic links: owner and group were
1439 5. Integrated some changes from FreeBSD 2.2.
1441 6. Added a call to ftruncate() after restoring each file to restore
1442 correctly files ending by a hole.
1444 Changes between versions 0.2d and 0.2e
1445 ======================================
1447 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1448 with owner = root (0) and group = root/wheel/whatever (0).
1450 Changes between versions 0.2c and 0.2d
1451 ======================================
1453 1. Dump is now able to backup 2GB+ filesystems.
1455 2. Dump and restore can now be linked as static binaries.
1457 Changes between versions 0.2b and 0.2c
1458 ======================================
1460 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1462 Changes between versions 0.2a and 0.2b
1463 ======================================
1465 1. Really fixed the bug that I should have corrected in 0.2a.
1467 2. Enabled optimization again.
1469 Changes between versions 0.2 and 0.2a
1470 =====================================
1472 1. Disabled the optimization during compilation.
1474 Changes between versions 0.1 and 0.2
1475 ====================================
1477 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1478 the fs_type field (actually, I modified the file fstab.c to make it
1479 use the mntent functions).
1481 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1482 made by dump should be readable by the BSD restore and backups made
1483 by the BSD dump should be readable by restore. Unfortunately, this
1484 means that the dump format has changed between version 0.1 and version
1487 3. Dump is now able to backup a subtree, it is no longer limited to whole
1488 filesystems like the BSD version.
1490 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1493 Changes between versions 0.0 and 0.1
1494 ====================================
1496 1. Now create links rdump and rrestore during the `make install' step.
1498 2. Linux port specific bugs added to the manual pages
1500 3. Incorrect estimation of the number of tapes blocks fixed when doing
1501 incremental backups.
1503 4. Better ls-like format in restore in interactive mode.
projects / dump.git / blob