1 $Id: CHANGES,v 1.276 2005/06/08 09:34:33 stelian Exp $
3 Changes between versions 0.4b40 and 0.4b41 (released ???????????)
4 =================================================================
6 1. Fix restore of dumped Access Control Lists. The previous
7 code for EA works fine, but ACLs needed conversion from the
8 ext2/3 disk format to posix_acl format before restoring.
10 Changes between versions 0.4b39 and 0.4b40 (released May 2, 2005)
11 =================================================================
13 1. Changed restore to emit warnings (instead of emitting a fatal
14 error) if a file (or a directory) is unavailable for a
15 comparision (if the user doesn't have the necessary permissions
16 to access it for example). Thanks to Kenneth Porter
17 <shiva@sewingwitch.com> for the bug report.
19 2. Re-done the 'do not save directory entries to non-dumped inodes
20 (excluded from dump)' feature. The previous implementation
21 worked well for excluded directories but not for regular files.
22 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
25 3. Fixed a bug in dump where the tape size was miscalculated when
26 the user used -d/-s to specify the tape characteristics. Thanks
27 to Philip Goisman <goisman@physics.arizona.edu> for reporting
30 4. Fixed another bug introduced in restore with the hashtree
31 implementation. This one caused restore to stop saying
32 "removenode: non-empty directory" in some cases.
34 5. Added support for dumping and restoring ext2/3 extended
35 attributes (EA), like the access control lists (ACL) or
36 the security labels used by SELinux.
38 Changes between versions 0.4b38 and 0.4b39 (released January 21, 2005)
39 ======================================================================
41 1. The newly added dump_on_cd_3 example was buggy, replace it
42 with an updated version from Andrew Basterfield
43 <bob@cemetery.homeunix.org>.
45 2. Made restore to chdir() back into the initial directory when
46 dumping core while aborting a comparision operation. The
47 previous behaviour was to write the corefile at the root of
48 the directory being compared, which could very well be
49 read only and preventing the corefile generation. Thanks
50 to Kenneth Porter <shiva@sewingwitch.com> for the bug report.
52 3. Silenced the failure to call fgetflags() when comparing an
53 entry which has no ext2 attributes (as in lsattr()).
55 4. Fix a brown paper bug in restore -C which broke restore and
56 caused modifications on the filesystem being compared
57 (directories containing a file with the same name as the
58 directory get renamed to RSTTMP...). Thanks to Kenneth Porter
59 <shiva@sewingwitch.com> for finding the bug and helping me
62 5. Made restore -C force the -N flag (no writing allowed on
63 the disk) in order to prevent more bugs like the above one.
65 Changes between versions 0.4b37 and 0.4b38 (released January 7, 2005)
66 =====================================================================
68 1. Fix a couple of troff syntax bugs in the man pages.
69 Thanks to Eric Raymond <esr@thyrsus.com> for the patch.
71 2. Made restore use either libncurses or libtermcap, depending
72 on which one is available at configure time.
74 3. Fixed restore negative size display bug when comparing a
75 dump containing files over 2GB. Thanks to Steve Bonds
76 <sbonds@users.sourceforge.net> for the bug report.
78 4. Do not save directory entries to non-dumped inodes
79 (excluded from dump). This will eliminate the 'missing
80 file' warnings when doing 'restore -C'.
82 5. Fix dump crash when backuping a huge (2TB) filesystem,
83 due to a bogus calculation on the inode map size.
84 Thanks to Kevin B. Haines <K.B.Haines@rl.ac.uk> for
85 submitting the bug and testing the fix.
87 6. Fix a problem in restore where the final \0 in the symbolic
88 link names could have been lost, generating corrupt filenames.
89 Thanks to Kyle Wilson <kyle.wilson@amd.com> for reporting the
92 7. Implemented a hash list for the directory names in restore.
93 The linear list used before caused problems in interactive
94 restores when dealing with directories having thousands of
95 entries. Thanks to Brian Ristuccia <bristuccia@starentnetworks.com>
96 for reporting the bug.
98 8. Improved restore -C, this time including the directory
99 attributes into the comparision.
101 9. Made restore understand tapes containing EA/ACLs (which will
102 be dumped by the next version of dump). In this version
103 extended attributes on the tape are ignored, for full EA/ACL
104 support wait for the next version or try the experimental EA
107 Changes between versions 0.4b36 and 0.4b37 (released July 7, 2004)
108 ==================================================================
110 1. Added the --enable-staticz configure option which enables
111 dump and restore to be linked with static versions of
112 libz and libbz2 (and dynamic versions of all other libraries).
113 This will make Debian users happy, because libz and libbz2
114 were the only needed libraries living in /usr, all the
115 others live in /lib. In case of system emergency, it is
116 better not to have to rely on an extra filesystem.
117 Thanks to Bdale Garbee <bdale@gag.com> for the suggestion.
119 2. Fix compilation on (at least the Linux Debian port to) AMD64.
120 (<ext2fs/ext2_types.h> defines some types (__s64 and __u64)
121 that are also defined by <linux/types.h> (<asm/types.h>) and
124 3. Make dump's reading of the dumpdates file a bit more robust,
125 preventing dump from crashing when the dumpdates file has
126 been modified by hand.
128 4. Fixed some offset calculations in dump code which could
129 lead to "bread lseek errors" on large filesystems. Thanks
130 to Bruce Lowekamp <lowekamp@users.sourceforge.net> for
131 reporting this bug and debugging the issue.
133 5. Made dump use the blkid library when searching for devices
134 by label or uuid instead of dump's own routines.
136 6. Corrected a bug in dump where a wrong LABEL=... line in
137 /etc/fstab could prevent dump from dumping unrelated
138 filesystems. Thanks to Bruce Lowekamp
139 <lowekamp@users.sourceforge.net> for reporting the bug.
141 Changes between versions 0.4b35 and 0.4b36 (released April 21, 2004)
142 ====================================================================
144 1. Fixed dump compilation with old gcc versions. Thanks to
145 Mike Castle <dalgoda@users.sourceforge.net> for the patch.
147 2. Fixed some warnings (howmany, roundup, powerof2 redefined)
148 when compiling against a recent glibc version.
150 3. Fixed a bug in restore preventing the read of a dump tape
151 written with Solaris 7 ufsdump. Thanks to Patrick Higgins
152 <phiggins@transzap.com> for reporting the bug and providing
155 4. Changed dump to enable the creation of volumes of different
156 sizes in a single run (make -B accept a list of values).
157 Patch contributed by Florian Zumbiehl <florz@gmx.de>.
159 5. Use the glibc provided minor() and major() macros instead
160 of our own bitmask implementation. This should be safe for
161 when the major/minor namespace will migrate to 32 bits.
162 Thanks to Zhang Jun <zhangjun@nanjing-fnst.com> for reporting
165 6. Made explicit in the dump man page that dump will not create
166 a remote file, it will only write to an already existing one.
168 7. Another try at making size estimates better again.
170 8. Put back the inconditional running of the end-of-tape
171 script which was changed in 0.434 to be run only when -M
172 or multiple -f were NOT specified. Some users rely on this
173 feature even when it is combined with -M/-f.
175 9. Fixed restore when restoring huge backups (where rstdir...
176 temporary files are over 2GB). Thanks to
177 Raphael Jaffey <rjaffey@artic.edu> for reporting this,
178 debugging the issue and making the patch.
180 10. Made restore understand FreeBSD UFS2 tapes. Thanks to
181 David <vrtin@users.sourceforge.net> for submitting the bug
182 and providing a test case.
184 11. Made dump work with an arbitrary integer level (no more 10
185 levels only limitation). Thanks to Matthew
186 <msvincen@midway.uchicago.edu> for the patch.
188 Changes between versions 0.4b34 and 0.4b35 (released December 21, 2003)
189 =======================================================================
191 1. Added a note in the dump man page saying that the default
192 blocksize can be 32 if -d is used with a high density argument.
193 Thanks to Antonios Christofides <A.Christofides@itia.ntua.gr>
196 2. Fixed configure to correctly understand CPPFLAGS, CFLAGS,
197 LDFLAGS environment variables. Thanks to Arcady Genkin
198 <antipode@users.sourceforge.net> for reporting the bug.
200 3. Made -e/-E options of dump accept an unlimited number of inodes
201 to be excluded, rather than a hardcoded maximum. Thanks to
202 Dietrich Rothe <d-rothe@users.sourceforge.net> for the patch.
204 4. Updated the autoconf system to 2.50. Forced the -D_BSD_SOURCE
205 and -D_USE_BSD_SIGNAL defines in configure in order to solve
206 64bit build problems because quad_t is redifined with a
207 different signature. Thanks to Mike Harris <mharris@redhat.com>
208 for reporting this bug.
210 5. Made restore build on Solaris, making possible to
211 restore Linux's "enhanced" tapes. Thanks to Uwe Gohlke
212 <uwe@ugsoft.de> for the patch.
214 6. Made an extension in the dump tape format capable of saving
215 MacOSX specific inode extensions. Uwe Gohlke <uwe@ugsoft.de>
216 wrote the extension and contributed the restore code back
217 into this codebase. The same extension mechanism will be
218 used in the future to save ACLs...
220 7. Made rmt work correctly with regard to QFA and local/remote
221 files and tapes. The remote access will however work only
222 when the dump provided rmt version is used. If you want to
223 use another rmt server, please do not use the QFA feature.
224 Thanks to Uwe Gohlke <uwe@ugsoft.de> for the patch.
226 Changes between versions 0.4b33 and 0.4b34 (released April 18, 2003)
227 ====================================================================
229 1. Fixed the INSTALL file to reflect the actual install paths.
230 Thanks to David Ronis <ronis@ronispc.chem.mcgill.ca> for
233 2. Fixed the configure script to only check for headers presence
234 instead of trying to compile them. This should fix issues
235 with old build environments. Thanks to Kari Lempiainen
236 <kari@funky.pp.fi> for reporting the bug.
238 3. Fixed restore to correctly ignore sockets when comparing
239 a dump (as socket cannot be properly restored anyway). Thanks
240 to Gunther Reiszig <gunther@mit.edu> for reporting the bug.
242 4. Fixed restore to correctly access the archive file (-A argument)
243 even when using a remote tape. Thanks to Fabrice Bellet
244 <fabrice@bellet.info> for reporting the bug.
246 5. Fixed (again) handling of long (largefile) seeks in rmt.
247 Thanks to Fabrice Bellet <fabrice@bellet.info> for reporting
250 6. Fixed restore corner case when dealing with large block sizes
251 dump is able to write now (-b 1024). Thanks to Fabrice Bellet
252 <fabrice@bellet.info> for reporting the bug.
254 7. Fixed a bug preventing dump to access a filesystem having
255 a label exactly 16 bytes in length. Thanks to <minduni@ti-edu.ch>
256 for reporting the bug.
258 8. Made dump store dump dates using explicit timezones, fixing a
259 problem with incremental dumps if the timezone is modified
260 between the dumps. Thanks to Philippe Troin <phil@fifi.org> for
261 the bug report and the patch.
263 9. Fixed a bug encountered when dumping individual files (not full
264 filesystems or directories) and dangling symbolic links happen
265 to be in the list of files. For as far as dump is concerned,
266 dangling symbolic links are allowed, and are dumped as is.
267 Thanks to Jin-su Ahn <jsahn@ee.snu.ac.kr> for reporting the
268 bug and providing the fix.
270 10. Fixed open and creation modes and permissions for QFA and
271 table-of-contents files in dump and restore. Thanks to
272 Philippe Troin <phil@fifi.org> for the patch.
274 11. Fixed the archive file descriptor handling enabling it to be 0.
275 This can happen in some cases when shell redirections are used.
276 Thanks to Philippe Troin <phil@fifi.org> for the patch.
278 12. Delayed the opening of archive file until after suid had been
279 dropped (fixing a possible security issue if dump is suid).
280 Thanks to Philippe Troin <phil@fifi.org> for the patch.
282 13. Fixed the 'S' command handling in the rmt client part.
283 Thanks to Philippe Troin <phil@fifi.org> for the patch.
285 14. Modified the end-of-tape script handling to print out statistics
286 (and stop the timer) before launching the eot script. Also, the eot
287 script does not get run anymore when using -M (which makes sense) or
288 when multiple tapes are listed on the command line
289 (-f tape0,tape1,tapen) (which also makes sense).
290 Thanks to Philippe Troin <phil@fifi.org> for the patch.
292 15. Relicensed dump/restore under the 'revised' BSD license, as per
293 ftp://ftp.cs.berkeley.edu/ucb/4bsd/README.Impt.License.Change.
295 16. Added LZO compression to dump. This new compression method has
296 the advantage of being super fast, thus not killing tape streaming
297 on slow machines. Thanks to Helmut Jarausch
298 <jarausch@igpm.rwth-aachen.de> for the patch and to
299 Markus Oberhumer <markus@oberhumer.com> for giving special permission
300 to include his miniLZO project (GPL licensed) in dump/restore.
302 17. Some small buffer overruns fixes in rmt. Thanks to Antonomasia
303 <ant@notatla.demon.co.uk> for reporting the bugs.
305 18. Added a special rmt version which can do encryption when writing
306 to tape. Read examples/encrypted_rmt/README for details on
307 how to enable and configure it. Thanks to Ken Lalonde
308 <ken@globalremit.com> for the patch.
310 19. Made dump work with 2.5 kernel end of tape early warning semantics.
311 Thanks to Kai Makisara <Kai.Makisara@kolumbus.fi> for the patch.
313 20. Fixed a bug which caused dump -w|-W not to work anymore, because
314 the fs_freq and fs_passno fields in /etc/mtab are always set
315 to 0 0. Thanks to Trent Piepho <xyzzy@speakeasy.org> for
318 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
319 =======================================================================
321 1. Added a note in the restore man page clarifying the question
322 'set the permissions on the current directory ?' asked by
323 restore at the end of treatment in -i and -x modes.
325 2. Fixed the endianess issues when reading compressed tapes.
326 Thanks to Dark Force <daq4th@users.sourceforge.net> for
327 reporting this bug and providing test cases.
329 3. Fixed the "ACL won't be dumped" warning message (which showed
330 an extra, unrelated error message). Thanks to Dragan Krnic
331 <dkrnic@lycos.com> for reporting this bug.
333 4. Made dump look first into /etc/mtab, then into /etc/fstab
334 when searching for filesystem paths. Also fixed some problems
335 caused by binding mounts or multiple block device mounts.
336 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
337 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
338 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
339 and #160305, Sourceforge bugs #618699 and #679832).
341 5. Made dump's -I option accept the value '0' meaning all the
342 read errors will be ignored. This can be useful when running
343 dump from unattended sessions (like cron jobs). Thanks to
344 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
346 6. Fixed the output of dump to indicate 'blocks' instead of
347 'tape blocks' in the various messages (blocks are always
348 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
349 made it clearly print the current blocksize at the start of
350 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
353 7. Made rmt understand long (largefiles) seeks.
355 8. Fixed build with very old versions of libext2fs, where
356 EXT2_FT_* constants were undefined.
358 9. Made dump accept the dumpdates path on the command line
359 (-D file option) instead of using only the hardcoded one.
360 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
363 10. Enabled rmt, LFS, readline, QFA options by default in
364 ./configure. Updated the configure process (new versions
365 of config.guess, config.sub etc).
367 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
368 =======================================================================
370 1. Changed dump to use fcntl(F_SETLK) style locking instead
371 of flock() when locking the dumpdates file. With the old
372 locking scheme, a local user having read rights on the
373 dumpdates file could be able to do a Denial of Service attack
374 on dump. In order to lock the dumpdates file with the new
375 scheme, the user would need to have write access on the file.
376 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
377 reporting the bug (originally a bugtraq post).
379 2. Fixed interactive 'ls' which caused spurious errors warnings
380 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
381 <jorgen_ostling@users.sourceforge.net> for reporting this
384 3. Fixed dump's estimate when dealing with sparse inodes.
386 4. Modified dump to allow setting a blocksize bigger than 32kB
387 (raised the limit to 1024kB), because newer hardware needs
388 this for proper operation. Thanks to Dirk Traenapp
389 <dtraenapp@users.sourceforge.net> for reporting this.
391 5. Fixed a bug causing Dump to stop and report an error if an
392 inode number in the exclude file was followed by some amount
393 of whitespace. Thanks to Jeffrey Sofferin
394 <sofferin@users.sourceforge.net> for reporting this bug.
396 6. Fixed a bug which caused restore, in some particular cases,
397 to ask some 'scary' questions and leave a bunch of RSTTMP
398 directories behind when restoring incremental tapes. Thanks
399 to Philippe Troin <phil@fifi.org> for reporting this bug and
400 providing the test cases.
402 7. Changed the wording when inodes are excluded from dump:
403 replaced 'Added inode 7 to exclude list' with
404 'Excluding inode 7 (resize inode) from dump', as suggested
405 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
407 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
408 ===================================================================
410 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
411 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
412 <eros@lamel.bo.cnr.it> for reporting the bug.
414 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
415 ===================================================================
417 1. Made dump print out the ext2 logical block number in case of a read
418 error, which can be given as an argument to debugfs' ncheck command,
419 in order to find out the affected inode. Added note about this
420 usage in dump's man page.
422 2. Fixed a problem in restore when reading tapes written on big
423 endian machines with very old versions of dump. The patch was
424 contributed by George Helffrich <george@geology.bristol.ac.uk>.
426 3. Fixed the tape length calculation when using large tapes
427 and compression. Thanks to Georg Lippold
428 <g_lippold@sourceforge.net> for reporting the bug.
430 4. Added a new set of examples in dump_on_cd_2 directory, based
431 on dump_on_cd examples but somewhat enhanced, supporting DVD
432 media, and localized in english and german. Thanks to
433 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
435 5. Made dump save 32 bit UID/GID. Internally, this was achieved
436 by switching from the old BSD inode format to the new BSD
437 inode format, which means that the tape format was changed.
438 However, since all restore versions out there should
439 transparently support both inode formats, the change should
440 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
443 6. Fixed a lot of warnings in the code shown when compiling
444 with 'gcc -W'. Thanks to Matthias Andree
445 <matthias.andree@stud.uni-dortmund.de> for reporting this.
447 7. Fixed a small markup bug in the dump man page. Thanks to
448 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
451 8. Rewrote entirely the man pages using the tmac.an macro
452 package (Linux man page format) instead of the original BSD
453 format. They should be now cleaner and easier to modify.
455 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
456 ==================================================================
458 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
459 non Linux clients were misinterpreted. The description of the problem
460 (incompatible numbering in Linux mtio opcodes) is documented at
461 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
462 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
463 providing an excellent, cross-platform replacement for rmt in his
466 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
467 which caused restore to fail when the inode maps were not entirely
468 contained on the first volume (when using really small volumes or
469 when dumping a huge number of inodes).
471 3. Fixed a problem in dump, where files in subdirectories of directories
472 that have the nodump flag set (or being excluded with -e/-E) were
473 sometimes incorrectly dumped (instead of being excluded). The original
474 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
475 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
476 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
477 help testing the patch.
479 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
480 which makes one able to pipe the output of dump, by the net, to
481 a remote CD-burner server.
483 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
484 and use GNU's symbolic syntax over rmt instead of numerical values
485 to assure multiple platform compatibility.
487 6. Documented the -d option in restore.
489 7. Added a -v (verbose) mode to dump. For now it just prints the number
490 of the inode being dumped, but this could evolve in future versions
491 to include interesting debugging output.
493 8. Added a -o flag to restore, which automatically restores the current
494 directory permissions in -i or -x mode, without asking the operator
495 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
496 and Peter Samuel <peters@e-smith.com>.
498 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
499 ====================================================================
501 1. Fixed a bug in the QFA feature which could have caused
502 a bad tape positionning on restore (causing extra delays in
503 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
505 2. Added a small note in the dump man page specifying that
506 there should be no white space between the option letter and
507 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
509 3. Made restore work with older versions of the readline library,
510 by checking in configure for several extended symbols. Restore
511 can now be compiled with a readline as old as the 2.0 release
512 (though it may be a good idea to upgrade it to a more recent
513 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
514 reporting the build failures.
516 4. Fixed a performance problem with the QFA file creation in
517 dump, which made unnecessary seeks on the tape slowing down
518 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
519 reporting this issue.
521 5. Removed the inclusion of some kernel headers in the dump
522 source, which prevented the compile in some kernel/glibc
523 headers/architecture combination. Thanks to Bdale Garbee
524 <bdale@gag.com> for reporting the bug.
526 6. Added the appropriate error message when dump fails to
527 open the output file for writing. Thanks to Amith Varghese
528 <amithv@yahoo.com> for reporting this bug.
530 7. Made restore able to understand large Solaris ufsdump tapes
531 (containing inodes bigger than 4194304). Sun have introduced
532 an "extension" to the dump tape format when dealing with
533 those inodes, which was uncorrectly handled by Linux restore.
534 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
535 providing a test case.
537 8. Added the -m parameter to dump which optimises the output for
538 inodes having been changed but not modified since the last dump
539 ('changed' and 'modified' have the meaning defined in stat(2)).
540 For those inodes, dump will save only the metadata, instead of
541 saving the entire inode contents. Inodes which are either
542 directories or have been modified since the last dump are saved
543 in a regular way. Uses of this flag must be consistent, meaning
544 that either every dump in an incremental dump set have the flag,
547 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
548 =======================================================================
550 1. Fixed behaviour of dump when exceeding resource limits
553 2. Added the -L flag to restore to allow the user to specify a
554 maximal allowed number of miscompares when using restore
555 with the -C option to check the backup.
557 3. Detailed the manual entry for the -N option of restore.
559 4. Added the -a flag to restore to make able doing unattended
560 restores in -i or -x mode (automatically walks through the
561 multiple dump volumes).
563 5. Extended the QFA mode to work with local files and/or
564 remote tapes and files. This way, restore can know in advance
565 the tape number and the offset for the inodes to extract and
566 can minimize the extraction time by seeking directly to the
569 6. Added the -A <archive> option to both dump and restore,
570 which makes dump to archive a dump table-of-contents in
571 the specified file to be used by restore to determine
572 whether a file is in the dump file that is being restored.
573 (the archive file syntax is also compatible with the
574 Solaris ufsdump generated one).
576 7. Small fix in restore making it able to read some (broken ?)
577 Solaris ufsdump tapes.
579 8. Fixed dump to correctly recognise the root filesystem when using
580 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
581 for reporting this bug.
583 9. Added the -P <file> option to restore to create a
584 Quick File Access file from an already made dump. Patch
585 contributed by Uwe Gohlke <uwe@ugsoft.de>.
587 10. Made restore compile and run on Solaris, making it a
588 possible replacement for the standard ufsrestore. Port was
589 contributed by Uwe Gohlke <uwe@ugsoft.de>.
591 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
592 =====================================================================
594 1. Added a set of backup scripts from Eugenio Diaz
595 <getnito@yahoo.com> in the examples section. It features
596 automatic (cron based) full and incremental dumping of
597 several filesystems on a separate filesystem.
599 2. Fixed a off-by-one miscalculation which disabled dumping
600 a one letter subdirectory. Thanks to Chris Siebenmann
601 <cks@utcc.utoronto.ca> for reporting the bug.
603 3. Fixed several restore bugs occuring when trying to
604 restore some missing files on the tape. Thanks to Chris
605 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
607 4. Fixed --with-ldopts configure argument passing, installing from
608 a separate object directory, makefile cleanups contributed
609 by <splite@purdue.edu>.
611 5. Fix a bug which could caused, in some conditions, the highest
612 number inode of a filesystem, to not be dumped. Many thanks
613 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
616 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
617 =======================================================================
619 1. Added a mini howto from Patrick Walsh in the examples
622 2. Minor man pages syntax corrections. Thanks to
623 Chris Danis <screechco@home.com> for reporting the bugs.
625 3. Added a script from David B. Peterson <dave@toppledwagon.com>
626 to the examples section. It features dumping several
627 filesystems to a remote tape drive upon ssh.
629 4. Added a patch provided by Richard Jones <rich@annexia.org>
630 which allows BRADEMAX (number of read errors tolerated by
631 dump) to be adjusted using the -I option.
633 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
634 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
637 6. Add the -l option to restore to specify if, when doing a
638 remote restore, the file used to access the backup is a
639 regular file (the defaults being a tape drive). Restore needs
640 to know this information when reading a remote compressed
641 dump. Previously, this information was autodetected, but
642 the autodetection code fails (with ioctl: Inappropriate ioctl
643 for device) when using a non Linux remote box. Thanks to
644 many users and especially to Eros Albertazzi
645 <eros@lamel.bo.cnr.it> for reporting this.
647 7. Found a workaroung for the dump deadlock problem (3 childs
648 stuck in pause(), father in read()). The workaround seems
649 to work for me and several beta-testers. If it doesn't work
650 for you, please report back.
652 8. Updated the RPM spec file (BuildPrereq, URL etc).
654 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
655 ========================================================================
657 1. Fixed the permissions of a newly created QFA file by dump.
659 2. Cleaned up the source of dump (the external variables
660 definition was a complete mess, making possible to have
663 3. Fixed restore to use the full tape volume path when doing
664 a compare (since it changes the working directory to the
665 filesystem being compared in the process).
667 4. Added the -q option to dump which makes dump abort
668 whenever it needs operator attention. It should be
669 easier to use dump in scripts with this option.
671 5. Detect the use of incompatible options to dump and
672 refuse them (like -a and -B options together).
674 6. Added bzip2 compression to dump/restore (use option -j level
675 to select it). Note that you will need a restore version
676 >= 0.4b24 in order to restore a bzip2 compressed dump.
677 The same warning as for the zlib compression applies:
678 the tape format of a bzip2 dump is not compatible with the
679 original BSD tape format.
681 7. Fixed a overflow problem in dump corrupting the dump when
682 very large files were encountered. Thanks to Vladimir Nadvornik
685 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
686 kernel buffer/page cache before starting the dump, helping
687 a bit those who use dump on mounted filesystems. Thanks to
688 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
691 9. Updated the RPM spec file following the RedHat changes
692 (dynamically linked binaries now in /usr/sbin etc).
694 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
695 which enables restore to recognise multi volume compressed dumps
696 done on CD. Included his scripts for dump (which pipe the dump
697 output directly into cdrecord) and restore. There is now
698 possible to to backups to CD on the fly!
700 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
701 ===================================================================
703 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
704 Marcus Meissner (Caldera International Security Dept.).
706 2. Implement the Sun rmt extended protocol. Patch submitted
707 by Ian Gordon <iangordon@users.sourceforge.net>.
709 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
710 the linux kernel header. This ensures that dump always has the
711 latest version of this file. Patch submitted by
712 Andreas Dilger <adilger@turbolinux.com>.
714 4. Report any filesystem present in either /etc/fstab with a
715 positive passno or /etc/dumpdates in dump -w output.
716 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
718 5. Fixed the looping problem in dump introduced in the
721 6. Changed the -B option of dump to limit the size of
722 _compressed_ output per volume if compression is on.
723 Patch contributed by Helmut Jarausch
724 <jarausch@igpm.rwth-aachen.de>. Note however that, since
725 it is impossible to predict the size of the compressed
726 data before doing the compression, the -B limit is a bit
729 7. Fixed a bug in reading the operator typed file/tape path for
730 the next volume in restore.
732 8. Implemented a "-F script" option to restore which permits the
733 user to specify a script which will be launched at the
734 beginning of each tape, useful for automatic programming of
735 tape changers for example. See the restore man page for the
736 script parameters and return codes.
738 9. Small fix for the QFA routines provided by Uwe Gohlke
739 <uwe@ugsoft.de>, and some recommendations for QFA uses in
742 10. Fixed the multivolume restoring where making a mistake
743 to the 'Mount next tape' prompt caused several blocks to
746 11. Enhanced the -e option of dump to take as a parameter a
747 comma separated list of inode numbers.
749 12. Added the -E option to dump which specify a file containing
750 inode numbers to exclude from the dump.
752 13. Fixed the compressed multi-volume dump + restore.
754 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
755 ==================================================================
757 1. Made dump report the number of blocks dumped per volume.
758 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
760 2. Fix a bug in dump where waiting too much at the 'change volumes'
761 question caused the volume to be rewritten. Thanks to
762 Graham Reed <greed@users.sourceforge.net> for reporting the
763 bug and providing a patch.
765 3. Added a compression option to dump, contributed by Jerry
766 Peters <gapeters@worldnet.att.net>.
768 WARNING: the compression works only on files/pipes and
769 drives supporting variable block size.
771 WARNING2: tapes created using the compression feature are
772 incompatible with the standard dump tape format, and a
773 version of restore >= 0.4b22 will be needed for extraction.
775 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
776 architectures. Thanks to Paul Slootman <paul@debian.org> for
777 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
780 5. Many cleanups (CPP defines, const char warnings, check of
781 ext2fs COMPAT flags, time_t cleanups, added libext2 version
782 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
784 6. Made --prefix option work in configure. All the install path
785 are now based on the configure parameters.
787 7. Added the Quick File Access mode in dump/restore, contributed
788 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
789 a file tape position for each inode, and this file is used by
790 restore (if called with parameter Q and the filename)
791 to directly position the tape at the file restore is currently
792 working on. This saves hours when restoring single files from
793 large backups, saves the tapes and the drive's head. Use
794 --enable-qfa option of configure to compile in the QFA support.
796 8. Added the possibility to dump several files and directories
797 in a single invocation of dump. Thanks to Uwe Gohlke
798 <uwe@ugsoft.de> for implementing this option.
800 9. Fixed the dumping and restoring of files with holes
801 (files ending with holes didn't get dumped or restored
804 10. Fixed a socket descriptor leak in dump, which leaved opened
805 3 file descriptors per dump process (and there is one dump
808 11. Fixed dump large file system support, by explicit use of
809 open64/lseek64/etc functions (explicit use needed because
810 e2fsprogs libraries don't behave well when compiled with
811 FILE_OFFSET_BITS=64).
813 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
814 ======================================================================
816 1. Fixed some bugs in the dump -w|-W logic introduced by
817 the previous version. Thanks to Andreas Dilger
818 <adilger@turbolinux.com> for his help on this one.
820 2. Fixed again a compilation problem when using old e2fs
821 headers (filesystem label related). Thanks to many users
822 who reported this stupid error.
824 3. Fixed a build problem on old lib5 based systems dealing with
825 _PATH_MNTTAB being undefined. Thanks to John Adams
826 <johna@onevista.com> for reporting the bug.
828 4. Improved the error detection in restore when creating
829 the temporary files in TMPDIR. Now restore will corectly
830 report a 'no space left on device' error instead of
831 strange errors which could imply an archive corruption.
832 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
833 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
835 5. Added the throughput information in dump's progression
836 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
839 6. Use libext2fs's inode scanning routines, which are particularly
840 robust when dealing with errors (especially when having some
841 bad blocks on the filesystem). Many thanks to Colin
842 <colin@horizon.com> for reporting the bug, investigating
843 the issues, finding a workaround, writing the patch and
844 fully testing it... (of course, if this patch will break
845 anything, he is to blame for :-)).
847 7. Made dump and restore LFS aware. Dump can dump a filesystem
848 containing large files, generate a large file on output and
849 restore can restore them. This feature is not enabled by
850 default, use --enable-largefile option of configure to enable
851 it (you need to have a LFS aware glibc though). Thanks to
852 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
853 and to Theodore T'so <tytso@valinux.com> for his always
856 8. Made dump ask upon a tape write error if it should rewrite
857 the current volume (assume this is a bad tape) or if it should
858 assume an end-of-tape condition (useful for tape drives which
859 behaves badly at the end of the tape). Thanks to Andreas
860 Dilger <adilger@turbolinux.com> for the suggestion.
862 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
863 =======================================================================
865 1. Fixed a small compilation problem due to a change
866 in the definintion of the struct sigaction in
867 glibc 2.0 / libc5. Thanks to Gunther Schlegel
868 <schlegel@riege.de> for reporting the bug and to
869 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
872 2. Modified the label and uuid analysis in order to be
873 self-contained (independant of kernel/libc headers). This
874 should fix the compile with older kernel/libc headers and
875 will preserve the functionality. Thanks to Bernhard Erdmann
876 <bernhard.erdmann@gmx.de> for reporting the bug.
878 3. The 'exclude inode' option, if applied to a directory
879 inode, excludes now correctly all the leaves of this
880 directory. Thanks to John R. Dennison
881 <gerdesas@users.sourceforge.net> for reporting the bug.
883 4. Fixed the '-e' option to disable the possibility
884 to exclude the root inode (which causes the dumps to
885 be unreadable by restore). Prevented array overflow
886 when multiple -e options are used.
888 5. Fixed dump to correctly interpret a filesystem argument
889 which is a mountpoint and it is not an absolute path
890 (as specified in the fstab). Thanks to Bernhard R. Erdmann
891 <be@berdmann.de> for reporting the bug.
893 6. Made dump able to backup files larger than 2 GB. Note that
894 dump still doesn't cope with files larger than 4 GB.
896 7. Restore the real uid and gid before invoking an external
897 RSH program (big hole when dump or restore is suid root!).
899 8. Get the values for the default tape device and the location
900 of fstab file from the system headers. Thanks to
901 Andreas Dilger <adilger@turbolinux.com> for the patch.
903 9. Made dump -w|-W report all recognized filesystems
904 present in either /etc/fstab or /etc/dumpdates, and present
905 the list in the same order as in fstab file. Thanks
906 to Andreas Dilger <adilger@turbolinux.com> for the patch.
908 10. Made dump's -a (automatic end of tape detection) the
909 default. Specifying one of -B, -c, -d or -s options will
910 override the EOM detection. Thanks to Andreas Dilger
911 <adilger@turbolinux.com> for the patch.
913 11. Save the ext2 filesystem label into the dump volume label.
914 Specifying a volume label on the command line (-L option)
915 will override this feature. Thanks to Andreas Dilger
916 <adilger@turbolinux.com> for the patch.
918 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
919 =====================================================================
921 1. Fixed the signal handling in dump (which I broke in 0.4b17)
922 which was causing several strange dump failures (dump
923 hanged or segmentation faults etc).
925 2. Specified the default blocksize in the dump man page.
927 3. Changed two info messages of restore to be written on stdout
928 instead of stderr in order to leave stderr only for errors
929 or warnings. Thanks to Stephen Atwell
930 <satwell@urbana.css.mot.com> for the suggestion.
932 4. Corrected an off by one calculation which prevented
933 dumping a directory having a 1 character name length.
934 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
935 for reporting the bug.
937 5. Reinforce dump in order to allow only level 0 and no
938 -u option when dumping a subdirectory, and document
939 this in the man page. Thanks to Bernhard Erdmann
940 <bernhard.erdmann@gmx.de> for reporting the bug.
942 6. Fixed a small harmless bug in restore which caused
943 in some conditions hard links to be restored several
944 times (therefore generation some warning messages).
945 Thanks to Kenneth Porter <shiva@well.com> for
948 7. Updated the RPM spec file to the latest RedHat version,
949 providing FHS packaging and other cosmetic changes.
950 You will need RPM version >= 3.0.5 in order to build the RPM.
952 8. Updated the configure script to check for libtermcap
953 before checking for libreadline, because we need this
954 library in order to compile the readline support.
956 9. Made dump understand the LABEL= and UUID= notation
957 both in /etc/fstab and on the command line. Note that
958 you will need the /proc filesystem in order to use
959 these notations. Thanks to Erik Troan <ewt@redhat.com>
960 for providing the patch.
962 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
963 ===================================================================
965 1. Fixed a potential buffer overflow in restore. Thanks
966 to Stan Bubrouski <satan@fastdial.net> for reporting
969 2. Fixed a readline-related bug which prevented
970 'cat DUMPFILE | restore -i -f -' from working. Thanks
971 to Charles Karney <karney@users.sourceforge.net>
974 3. Changed a few "panic" into "exit", causing restore to
975 be more stable against some attacks (like the last one
976 reported on Bugtraq, although the last version of restore
977 was not vulnerable - just dumped core). Thanks to
978 Andreas Hasenack <andreas@conectiva.com.br> for reporting
981 4. Removed the suid-root bit on dump and restore in the
982 default build (and generated RPMs). It should be safer
983 now. Users who need the suid-root capabilities in order
984 to make network backups should read first the man page
985 of dump and enable the suid bit manually.
987 5. Added -ltermcap to the compile parameters for restore
988 when using readline support, in order to make the compile
989 process work on some systems (SuSE ?). Thanks to
990 Patrik Schindler <poc@pocnet.net> for reporting the bug.
992 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
993 ====================================================================
995 1. The -F script is called now *only* at the end of a tape,
996 not at the end of the dump. Thanks to Christian Haul
997 <haul@informatik.tu-darmstadt.de> for the bug report.
999 Normally, the device name and the current volume number
1000 are passed on the command line of the script. However,
1001 if you want the old style script (with no arguments
1002 passed on the command line), you can enable it in
1003 configure with the --enable-oldstylefscript.
1005 2. Use posix signal handling to preserve dump functionality
1006 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
1009 3. Made the exit code of restore in case of a 'C'ompare
1010 command reflect the result of the compare. An exit status
1011 of 0 means the dump archive is correct, 1 means tape errors,
1012 2 means that some files were modified. Thanks to Kenneth Porter
1013 <shiva@well.com> for the suggestion.
1015 4. Made (finally) quotes work in interactive restore.
1017 5. Small fixes in order to allow dump to compile with a
1018 really old e2fsprogs version. Thanks to Ian Zimmerman
1019 <itz@speakeasy.org> for the bug report.
1021 6. Add GNU readline capabilities to interactive restore.
1022 Use configure's --enable-readline flag to enable this feature.
1023 Thanks to Patrik Schindler <poc@pocnet.net> for the
1026 7. Do the compare on the fly in restore 'C' mode (this will
1027 allow not to exhaust the available /tmp space when
1028 ccmparing large files). Thanks to Kenneth Porter
1029 <shiva@well.com> for the suggestion.
1031 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
1032 ====================================================================
1034 1. Fixed some several duplicate 'DUMP: DUMP:' in the
1037 2. Corrected the estimation of blocks to dump. Note that
1038 this is still wrong for large directory inodes, due
1039 to the size of a BSD directory entry on the tape
1040 (estimation is based on the size of the ext2 directory,
1041 which is smaller because it doesn't contain the null
1042 character at the end).
1044 3. Made dump report the total number of MB written to
1045 tape at the end of run. Thanks to W. Reilly Cooley
1046 <wcooley@nakedape.cc> for the patch.
1048 4. Added the -X option to restore giving the possibility
1049 to read the names of the files to be extracted/listed
1050 from a text file (in addition of the command line).
1051 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
1054 5. Added the device name and the current volume number
1055 as arguments to the end of tape script (dump -F option).
1057 6. Made the multi-volume dump work again (after having
1058 broken it in 0.4b15).
1060 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
1061 ===================================================================
1063 1. Added a prompt command in interactive restore mode. Thanks
1064 to Andreas Dilger <adilger@home.com> for the patch.
1066 2. Fixed a buffer overflow problem in dump (caused by
1067 not checking the size of the filesystem parameter).
1068 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
1069 reporting this on Bugtraq (and to several dump users
1070 who forwarded me his mail).
1072 3. Added the '-F script' option to dump in order to
1073 launch a script at the end of each tape (to be used
1074 with a tape changer, or to notify the sysadmin by
1077 4. Fixed a bug in restore compare code caused by the changes
1080 5. Fixed the treatment of options using the old BSD syntax
1081 in both dump and restore.
1083 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
1084 =======================================================================
1086 1. Fixed a bug in dump which may have caused invalid deleted
1087 directories to be dumped out if they were deleted after the
1088 mapping pass. This could occure on active filesystem and lead
1089 to heap corruption (causing dump malfunction in many possible ways).
1090 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
1091 bug and submitting the patch.
1093 2. Fixed the handling of the filesystem parameter in dump. A
1094 '/mnt/disk' parameter caused the disk contents to be dumped,
1095 but a '/mnt/disk/' parameter caused the mountpoint directory
1096 to be dumped (generally an empty directory).
1098 3. Improved the output of dump in order to tell which directory
1099 it is currently dumping (when dumping a subtree).
1101 4. Added the '-e' exclude inode option to dump. Thanks to
1102 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
1104 5. Added a REPORTING-BUGS file in order to provide a guide
1105 on how to correctly report a bug in dump/restore.
1107 6. Fixed a restore problem when restoring a hard link to an inode
1108 having the immutable or append-only attribute set. Thanks to
1109 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
1111 7. Fixed a compatibility problem between dumps made on little
1112 endian machines (the format was correct) and big endian
1113 machines (the format was incorrect). This fix break the
1114 compatibility with the older dumps made on big endian
1115 machines (sparc, m86k, ppc etc). For the first time in
1116 linux dump's history (I believe), the dumps made by *BSD,
1117 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
1118 so interchangeable. Thanks to Rob Cermak
1119 <cermak@ahab.rutgers.edu> for submitting the bug and
1120 helping me test the fix.
1122 8. Fixed the way dump reports the remaining percent/time, if
1123 the number of blocks actually dumped exceeds the estimated
1124 number of blocks. Thanks to Jean-Paul van der Jagt
1125 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
1127 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
1128 ======================================================================
1130 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
1131 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
1133 2. Removed the requirement to build the RPM as root from the
1134 spec file. Thanks to Christian Weisgerber
1135 <naddy@mips.rhein-neckar.de> for submitting this.
1137 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
1138 causing dump to block when dumping really huge filesystems.
1139 Many thanks to Patrik Schindler <poc@pocnet.net> for
1140 helping me find this bug.
1142 4. Fixed the treatment for an interrupt signal when dump access
1143 the remote tape through RSH. Thanks to Christian Weisgerber
1144 <naddy@mips.rhein-neckar.de> for providing the patch.
1146 5. Fixed a bug which was causing dump/restore to display
1147 garbage characters instead of the remote host name.
1149 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
1150 =====================================================================
1152 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
1153 <kukuk@suse.de> for submitting the patch.
1155 2. Fix for the exit code when using the size estimate option of
1156 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
1157 submitting the patch.
1159 3. Handle EINTR in atomical reads/writes in dump, which was causing
1160 dump to fail on some systems. Thanks to Eric Jergensen
1161 <eric@dvns.com> for reporting the bug and submitting the patch.
1163 4. Handle more than 16 characters for the device names in dumpdates.
1164 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
1165 tracking down the problem and proposing the solution.
1167 5. Fixed a bug in dump which prevented the creation of the
1168 dumpdates file when doing a 0-level dump without already
1169 having a dumpdates file. Thanks to Patrik Schindler
1170 <poc@pocnet.net> for reporting the bug.
1172 6. Changed the way dump 'S' flag reports the size estimate
1173 from number of blocks into bytes (making it compatible
1174 with the Solaris version, and simplifying things for
1175 amanda users). Thanks to Jason L Tibbitts III
1176 <tibbs@math.uh.edu> for reporting the bug.
1178 7. Fixed a compatibility problem in linux/alpha dump tape format.
1179 Now the linux/alpha dump are (again) compatible with the
1180 other dump formats. But this breaks compatibility with
1181 older dumps made on alpha. Thanks to Mike Tibor
1182 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
1184 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
1185 ======================================================================
1187 1. Added a '--enable-kerberos' to configure.
1189 2. Added a 'S' option to dump which determines the amount of space
1190 that is needed to perform the dump without actually doing it, similar
1191 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
1192 <cermak@ahab.rutgers.edu>.
1194 3. Added a 'M' multi-volume option to dump and restore which enables
1195 dumping to multiple files (useful when dumping to an ext2
1196 partition to make several dump files in order to bypass the 2GB
1197 file size limitation). The 'f' argument is treated as a prefix and
1198 the output files will be named <prefix>001, <prefix>002 etc. With
1199 the 'M' flag, restore automatically selects the right file without
1200 asking to enter a new tape each time.
1202 4. Fixed a memory leak which was causing dump to grow very big
1203 (270MB when dumping a 10GB filesystem...). Thanks to Jason
1204 Fearon <jasonf@netrider.org.au> for reporting the bug.
1206 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
1207 ======================================================================
1209 1. Make configure test if the system glob routines support
1210 extended syntax (ALTDIRFUNC). If not, use the internal glob
1211 routines instead of system ones. Thanks to Bernhard Sadlowski
1212 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
1213 and helping me resolve this and other minor libc5 compiling
1216 2. Fix a problem when dumping a ext2fs with the 'filetype'
1217 feature enabled. Thanks to Patrick J. LoPresti
1218 <patl@cag.lcs.mit.edu> for reporting the bug and to
1219 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
1221 3. Made the nodump flag work on directories. A directory which
1222 has the nodump flag gets never dumped, regardless of its
1225 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
1226 which allows dump on an active ext3 filesystem. However, this
1227 is a "quick and dirty" patch which enables backup of an ext3
1228 filesystem through the ext2 compatibility (by ignoring the
1229 NEEDS_RECOVERY bit). The journal file is not recognized and
1230 it is dumped (it should not).
1232 5. Test the superblock compatibility flags when dumping, in order
1233 to be sure that we know how to deal with specific features.
1235 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
1236 ====================================================================
1238 1. Use lchown instead of chown, fixing a possible security problem
1239 when restoring symlinks (a malicious user could use this
1240 to deliberately corrupt the ownership of important system files).
1241 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
1242 this and providing the patch.
1244 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
1245 ====================================================================
1247 1. Put dump sources under CVS, added Id tags in all files so
1248 one can use 'ident' on binary files.
1250 2. Added the dump/restore version in the usage text so one can
1251 easily verify the version he is using.
1253 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
1254 a va_start/va_end problem on linux-ppc (always call va_start
1255 va_end in pairs each time we use a vararg function).
1257 4. Added again the DT_* constants because old libc does not
1258 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
1259 for submitting the bug report.
1261 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
1262 this should enable dumping big (huge) filesystems.
1264 6. Added the RSH environment variable in order to be able to
1265 use a rsh replacement like ssh when doing remote backups (and
1266 bypass the security limitations of rcmd). Now you can do remote
1267 backups without being root (or making dump setuid root).
1269 7. Modified again the way dumpdates works. For incremental dumps,
1270 we need to read dumpdates even if we are not using 'u' option.
1271 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
1274 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
1275 ===================================================================
1277 1. Removed the 'k' flag from the restore 'about' text if kerberos
1278 was not compiled in.
1280 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
1281 to them (fsetflags takes a char*, setflags an open fd!).
1283 3. (f)setflags is called only if the flags aren't empty. If the
1284 file is a special file, a warning is printed, because changing
1285 flags implies opening the device. Normally, a special file
1286 should not have any flag... (Debian bug #29775, patch provided
1287 by Abhijit Dasgupta <abhijit@ans.net>).
1289 4. Made possible to dump a file system not mentioned in /etc/fstab.
1290 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
1292 5. Changed the default behaviour to not create dumpdates
1293 unless 'u' option is specified. Removed the old "debian-patch"
1294 which provided the same thing. (Debian bug #38136, #33818).
1296 6. Removed all those dump*announce, since they were getting old...
1298 7. Added warning messages if dumpdates does not exist and
1299 when an inferior level dump does not exist (except for a level 0
1302 8. Debugged the glob calls in interactive mode: restore used a
1303 dirent struct which was different from the /usr/include/dirent.h
1304 one (this used to work, is it a glibc2 change?), so none of the
1305 compat glob (which used /usr/include/dirent.h) or the system glob
1306 worked. Restore use now the system dirent (and the system
1307 DT_* constants), which are compatible with BSD ones.
1309 9. Added a configure flag (--with-dumpdatespath) to specify
1310 the location of dumpdates file. By default, it is
1313 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1314 included the current date/version in man pages.
1316 11. Corrected the estimation of remaining time when
1317 the operator doesn't change the tapes quickly enough. This
1318 was an old bug, I thought I corrected it, and discovered
1319 that in fact it was corrected in two different places, so
1320 the results canceled each other...
1322 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1323 ===================================================================
1325 1. Integrated multiple patches from RedHat, Debian and SuSE:
1327 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1329 - several fixes in the man pages.
1330 - update the default tape device to /dev/st0.
1331 - many updates for Linux Alpha (byte ordering, size_t etc).
1333 - use environment variable for TMPDIR (instead of /tmp).
1334 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1335 - workaround egcs bug (RedHat bugs #4281 and #2989).
1336 - wire $(OPT) throughout Makefile's.
1338 2. Upgrade the dump revision to 1, making possible to dump filesystems
1339 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1341 3. Fix some compile warnings, prototype all functions.
1343 4. Use glibc err/glob instead of internal compatibility
1344 routines (only if available).
1346 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1347 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1349 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1350 ======================================================================
1352 1. Integrated the changes from FreeBSD-3.1-RELEASE
1353 (mostly bug fixes, buffer overruns, dump has now an "automatic
1354 tape length calculation" flag, dump/restore can use kerberos now
1355 (this is NOT tested), use environment variables for TAPE and
1358 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1360 3. Corrected a bug in restore when using 'C' option with multi-volumes
1361 tapes (files splited accros two tapes give "size changed" errors
1364 4. Corrected the long standing bug when dumping multiple tapes.
1365 This works for me, needs further testing.
1367 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1368 ====================================================================
1370 1. Dump now runs correctly on kernels 2.1.x
1371 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1373 Changes between versions 0.4b2 and 0.4b3
1374 ========================================
1376 1. Use realpath() if available
1378 2. Report statistics
1380 Changes between versions 0.4b1 and 0.4b2
1381 ========================================
1383 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1386 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1388 3. Fixed configure.in to correctly handle --enable-rmt
1390 Changes between versions 0.3 and 0.4b1
1391 ======================================
1393 1. Integrated the changes from 4.4BSD-Lite2
1395 2. Integrated the patches from Debian and Red Hat
1397 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1399 4. Changed dump to use the Ext2fs library to get block addresses. This
1400 should solve the endianness problem on SparcLinux.
1402 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1403 distribution's one) to use autoconf
1405 6. Fixed a few minor bugs
1407 Changes between versions 0.2e and 0.2f
1408 ======================================
1410 1. Added the creation of named pipes (fifos) in restore.
1412 2. Added the -N flag in the restore manual page.
1414 3. Added the file kernel.patch which contains the llseek() optimization
1415 patch for 1.2.x kernels.
1417 4. Fixed a bug in the restoration of symbolic links: owner and group were
1420 5. Integrated some changes from FreeBSD 2.2.
1422 6. Added a call to ftruncate() after restoring each file to restore
1423 correctly files ending by a hole.
1425 Changes between versions 0.2d and 0.2e
1426 ======================================
1428 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1429 with owner = root (0) and group = root/wheel/whatever (0).
1431 Changes between versions 0.2c and 0.2d
1432 ======================================
1434 1. Dump is now able to backup 2GB+ filesystems.
1436 2. Dump and restore can now be linked as static binaries.
1438 Changes between versions 0.2b and 0.2c
1439 ======================================
1441 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1443 Changes between versions 0.2a and 0.2b
1444 ======================================
1446 1. Really fixed the bug that I should have corrected in 0.2a.
1448 2. Enabled optimization again.
1450 Changes between versions 0.2 and 0.2a
1451 =====================================
1453 1. Disabled the optimization during compilation.
1455 Changes between versions 0.1 and 0.2
1456 ====================================
1458 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1459 the fs_type field (actually, I modified the file fstab.c to make it
1460 use the mntent functions).
1462 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1463 made by dump should be readable by the BSD restore and backups made
1464 by the BSD dump should be readable by restore. Unfortunately, this
1465 means that the dump format has changed between version 0.1 and version
1468 3. Dump is now able to backup a subtree, it is no longer limited to whole
1469 filesystems like the BSD version.
1471 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1474 Changes between versions 0.0 and 0.1
1475 ====================================
1477 1. Now create links rdump and rrestore during the `make install' step.
1479 2. Linux port specific bugs added to the manual pages
1481 3. Incorrect estimation of the number of tapes blocks fixed when doing
1482 incremental backups.
1484 4. Better ls-like format in restore in interactive mode.
projects / dump.git / blob