1 $Id: CHANGES,v 1.279 2005/07/07 09:16:03 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 2. Fix some issues with restoration of EA on big endian
13 3. Fix restore when the symtab is over 2GB in size.
15 4. Made the directory hash indexing an optional feature,
16 accessible by the '-H' option of restore, and disable it
19 Changes between versions 0.4b39 and 0.4b40 (released May 2, 2005)
20 =================================================================
22 1. Changed restore to emit warnings (instead of emitting a fatal
23 error) if a file (or a directory) is unavailable for a
24 comparision (if the user doesn't have the necessary permissions
25 to access it for example). Thanks to Kenneth Porter
26 <shiva@sewingwitch.com> for the bug report.
28 2. Re-done the 'do not save directory entries to non-dumped inodes
29 (excluded from dump)' feature. The previous implementation
30 worked well for excluded directories but not for regular files.
31 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
34 3. Fixed a bug in dump where the tape size was miscalculated when
35 the user used -d/-s to specify the tape characteristics. Thanks
36 to Philip Goisman <goisman@physics.arizona.edu> for reporting
39 4. Fixed another bug introduced in restore with the hashtree
40 implementation. This one caused restore to stop saying
41 "removenode: non-empty directory" in some cases.
43 5. Added support for dumping and restoring ext2/3 extended
44 attributes (EA), like the access control lists (ACL) or
45 the security labels used by SELinux.
47 Changes between versions 0.4b38 and 0.4b39 (released January 21, 2005)
48 ======================================================================
50 1. The newly added dump_on_cd_3 example was buggy, replace it
51 with an updated version from Andrew Basterfield
52 <bob@cemetery.homeunix.org>.
54 2. Made restore to chdir() back into the initial directory when
55 dumping core while aborting a comparision operation. The
56 previous behaviour was to write the corefile at the root of
57 the directory being compared, which could very well be
58 read only and preventing the corefile generation. Thanks
59 to Kenneth Porter <shiva@sewingwitch.com> for the bug report.
61 3. Silenced the failure to call fgetflags() when comparing an
62 entry which has no ext2 attributes (as in lsattr()).
64 4. Fix a brown paper bug in restore -C which broke restore and
65 caused modifications on the filesystem being compared
66 (directories containing a file with the same name as the
67 directory get renamed to RSTTMP...). Thanks to Kenneth Porter
68 <shiva@sewingwitch.com> for finding the bug and helping me
71 5. Made restore -C force the -N flag (no writing allowed on
72 the disk) in order to prevent more bugs like the above one.
74 Changes between versions 0.4b37 and 0.4b38 (released January 7, 2005)
75 =====================================================================
77 1. Fix a couple of troff syntax bugs in the man pages.
78 Thanks to Eric Raymond <esr@thyrsus.com> for the patch.
80 2. Made restore use either libncurses or libtermcap, depending
81 on which one is available at configure time.
83 3. Fixed restore negative size display bug when comparing a
84 dump containing files over 2GB. Thanks to Steve Bonds
85 <sbonds@users.sourceforge.net> for the bug report.
87 4. Do not save directory entries to non-dumped inodes
88 (excluded from dump). This will eliminate the 'missing
89 file' warnings when doing 'restore -C'.
91 5. Fix dump crash when backuping a huge (2TB) filesystem,
92 due to a bogus calculation on the inode map size.
93 Thanks to Kevin B. Haines <K.B.Haines@rl.ac.uk> for
94 submitting the bug and testing the fix.
96 6. Fix a problem in restore where the final \0 in the symbolic
97 link names could have been lost, generating corrupt filenames.
98 Thanks to Kyle Wilson <kyle.wilson@amd.com> for reporting the
101 7. Implemented a hash list for the directory names in restore.
102 The linear list used before caused problems in interactive
103 restores when dealing with directories having thousands of
104 entries. Thanks to Brian Ristuccia <bristuccia@starentnetworks.com>
105 for reporting the bug.
107 8. Improved restore -C, this time including the directory
108 attributes into the comparision.
110 9. Made restore understand tapes containing EA/ACLs (which will
111 be dumped by the next version of dump). In this version
112 extended attributes on the tape are ignored, for full EA/ACL
113 support wait for the next version or try the experimental EA
116 Changes between versions 0.4b36 and 0.4b37 (released July 7, 2004)
117 ==================================================================
119 1. Added the --enable-staticz configure option which enables
120 dump and restore to be linked with static versions of
121 libz and libbz2 (and dynamic versions of all other libraries).
122 This will make Debian users happy, because libz and libbz2
123 were the only needed libraries living in /usr, all the
124 others live in /lib. In case of system emergency, it is
125 better not to have to rely on an extra filesystem.
126 Thanks to Bdale Garbee <bdale@gag.com> for the suggestion.
128 2. Fix compilation on (at least the Linux Debian port to) AMD64.
129 (<ext2fs/ext2_types.h> defines some types (__s64 and __u64)
130 that are also defined by <linux/types.h> (<asm/types.h>) and
133 3. Make dump's reading of the dumpdates file a bit more robust,
134 preventing dump from crashing when the dumpdates file has
135 been modified by hand.
137 4. Fixed some offset calculations in dump code which could
138 lead to "bread lseek errors" on large filesystems. Thanks
139 to Bruce Lowekamp <lowekamp@users.sourceforge.net> for
140 reporting this bug and debugging the issue.
142 5. Made dump use the blkid library when searching for devices
143 by label or uuid instead of dump's own routines.
145 6. Corrected a bug in dump where a wrong LABEL=... line in
146 /etc/fstab could prevent dump from dumping unrelated
147 filesystems. Thanks to Bruce Lowekamp
148 <lowekamp@users.sourceforge.net> for reporting the bug.
150 Changes between versions 0.4b35 and 0.4b36 (released April 21, 2004)
151 ====================================================================
153 1. Fixed dump compilation with old gcc versions. Thanks to
154 Mike Castle <dalgoda@users.sourceforge.net> for the patch.
156 2. Fixed some warnings (howmany, roundup, powerof2 redefined)
157 when compiling against a recent glibc version.
159 3. Fixed a bug in restore preventing the read of a dump tape
160 written with Solaris 7 ufsdump. Thanks to Patrick Higgins
161 <phiggins@transzap.com> for reporting the bug and providing
164 4. Changed dump to enable the creation of volumes of different
165 sizes in a single run (make -B accept a list of values).
166 Patch contributed by Florian Zumbiehl <florz@gmx.de>.
168 5. Use the glibc provided minor() and major() macros instead
169 of our own bitmask implementation. This should be safe for
170 when the major/minor namespace will migrate to 32 bits.
171 Thanks to Zhang Jun <zhangjun@nanjing-fnst.com> for reporting
174 6. Made explicit in the dump man page that dump will not create
175 a remote file, it will only write to an already existing one.
177 7. Another try at making size estimates better again.
179 8. Put back the inconditional running of the end-of-tape
180 script which was changed in 0.434 to be run only when -M
181 or multiple -f were NOT specified. Some users rely on this
182 feature even when it is combined with -M/-f.
184 9. Fixed restore when restoring huge backups (where rstdir...
185 temporary files are over 2GB). Thanks to
186 Raphael Jaffey <rjaffey@artic.edu> for reporting this,
187 debugging the issue and making the patch.
189 10. Made restore understand FreeBSD UFS2 tapes. Thanks to
190 David <vrtin@users.sourceforge.net> for submitting the bug
191 and providing a test case.
193 11. Made dump work with an arbitrary integer level (no more 10
194 levels only limitation). Thanks to Matthew
195 <msvincen@midway.uchicago.edu> for the patch.
197 Changes between versions 0.4b34 and 0.4b35 (released December 21, 2003)
198 =======================================================================
200 1. Added a note in the dump man page saying that the default
201 blocksize can be 32 if -d is used with a high density argument.
202 Thanks to Antonios Christofides <A.Christofides@itia.ntua.gr>
205 2. Fixed configure to correctly understand CPPFLAGS, CFLAGS,
206 LDFLAGS environment variables. Thanks to Arcady Genkin
207 <antipode@users.sourceforge.net> for reporting the bug.
209 3. Made -e/-E options of dump accept an unlimited number of inodes
210 to be excluded, rather than a hardcoded maximum. Thanks to
211 Dietrich Rothe <d-rothe@users.sourceforge.net> for the patch.
213 4. Updated the autoconf system to 2.50. Forced the -D_BSD_SOURCE
214 and -D_USE_BSD_SIGNAL defines in configure in order to solve
215 64bit build problems because quad_t is redifined with a
216 different signature. Thanks to Mike Harris <mharris@redhat.com>
217 for reporting this bug.
219 5. Made restore build on Solaris, making possible to
220 restore Linux's "enhanced" tapes. Thanks to Uwe Gohlke
221 <uwe@ugsoft.de> for the patch.
223 6. Made an extension in the dump tape format capable of saving
224 MacOSX specific inode extensions. Uwe Gohlke <uwe@ugsoft.de>
225 wrote the extension and contributed the restore code back
226 into this codebase. The same extension mechanism will be
227 used in the future to save ACLs...
229 7. Made rmt work correctly with regard to QFA and local/remote
230 files and tapes. The remote access will however work only
231 when the dump provided rmt version is used. If you want to
232 use another rmt server, please do not use the QFA feature.
233 Thanks to Uwe Gohlke <uwe@ugsoft.de> for the patch.
235 Changes between versions 0.4b33 and 0.4b34 (released April 18, 2003)
236 ====================================================================
238 1. Fixed the INSTALL file to reflect the actual install paths.
239 Thanks to David Ronis <ronis@ronispc.chem.mcgill.ca> for
242 2. Fixed the configure script to only check for headers presence
243 instead of trying to compile them. This should fix issues
244 with old build environments. Thanks to Kari Lempiainen
245 <kari@funky.pp.fi> for reporting the bug.
247 3. Fixed restore to correctly ignore sockets when comparing
248 a dump (as socket cannot be properly restored anyway). Thanks
249 to Gunther Reiszig <gunther@mit.edu> for reporting the bug.
251 4. Fixed restore to correctly access the archive file (-A argument)
252 even when using a remote tape. Thanks to Fabrice Bellet
253 <fabrice@bellet.info> for reporting the bug.
255 5. Fixed (again) handling of long (largefile) seeks in rmt.
256 Thanks to Fabrice Bellet <fabrice@bellet.info> for reporting
259 6. Fixed restore corner case when dealing with large block sizes
260 dump is able to write now (-b 1024). Thanks to Fabrice Bellet
261 <fabrice@bellet.info> for reporting the bug.
263 7. Fixed a bug preventing dump to access a filesystem having
264 a label exactly 16 bytes in length. Thanks to <minduni@ti-edu.ch>
265 for reporting the bug.
267 8. Made dump store dump dates using explicit timezones, fixing a
268 problem with incremental dumps if the timezone is modified
269 between the dumps. Thanks to Philippe Troin <phil@fifi.org> for
270 the bug report and the patch.
272 9. Fixed a bug encountered when dumping individual files (not full
273 filesystems or directories) and dangling symbolic links happen
274 to be in the list of files. For as far as dump is concerned,
275 dangling symbolic links are allowed, and are dumped as is.
276 Thanks to Jin-su Ahn <jsahn@ee.snu.ac.kr> for reporting the
277 bug and providing the fix.
279 10. Fixed open and creation modes and permissions for QFA and
280 table-of-contents files in dump and restore. Thanks to
281 Philippe Troin <phil@fifi.org> for the patch.
283 11. Fixed the archive file descriptor handling enabling it to be 0.
284 This can happen in some cases when shell redirections are used.
285 Thanks to Philippe Troin <phil@fifi.org> for the patch.
287 12. Delayed the opening of archive file until after suid had been
288 dropped (fixing a possible security issue if dump is suid).
289 Thanks to Philippe Troin <phil@fifi.org> for the patch.
291 13. Fixed the 'S' command handling in the rmt client part.
292 Thanks to Philippe Troin <phil@fifi.org> for the patch.
294 14. Modified the end-of-tape script handling to print out statistics
295 (and stop the timer) before launching the eot script. Also, the eot
296 script does not get run anymore when using -M (which makes sense) or
297 when multiple tapes are listed on the command line
298 (-f tape0,tape1,tapen) (which also makes sense).
299 Thanks to Philippe Troin <phil@fifi.org> for the patch.
301 15. Relicensed dump/restore under the 'revised' BSD license, as per
302 ftp://ftp.cs.berkeley.edu/ucb/4bsd/README.Impt.License.Change.
304 16. Added LZO compression to dump. This new compression method has
305 the advantage of being super fast, thus not killing tape streaming
306 on slow machines. Thanks to Helmut Jarausch
307 <jarausch@igpm.rwth-aachen.de> for the patch and to
308 Markus Oberhumer <markus@oberhumer.com> for giving special permission
309 to include his miniLZO project (GPL licensed) in dump/restore.
311 17. Some small buffer overruns fixes in rmt. Thanks to Antonomasia
312 <ant@notatla.demon.co.uk> for reporting the bugs.
314 18. Added a special rmt version which can do encryption when writing
315 to tape. Read examples/encrypted_rmt/README for details on
316 how to enable and configure it. Thanks to Ken Lalonde
317 <ken@globalremit.com> for the patch.
319 19. Made dump work with 2.5 kernel end of tape early warning semantics.
320 Thanks to Kai Makisara <Kai.Makisara@kolumbus.fi> for the patch.
322 20. Fixed a bug which caused dump -w|-W not to work anymore, because
323 the fs_freq and fs_passno fields in /etc/mtab are always set
324 to 0 0. Thanks to Trent Piepho <xyzzy@speakeasy.org> for
327 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
328 =======================================================================
330 1. Added a note in the restore man page clarifying the question
331 'set the permissions on the current directory ?' asked by
332 restore at the end of treatment in -i and -x modes.
334 2. Fixed the endianess issues when reading compressed tapes.
335 Thanks to Dark Force <daq4th@users.sourceforge.net> for
336 reporting this bug and providing test cases.
338 3. Fixed the "ACL won't be dumped" warning message (which showed
339 an extra, unrelated error message). Thanks to Dragan Krnic
340 <dkrnic@lycos.com> for reporting this bug.
342 4. Made dump look first into /etc/mtab, then into /etc/fstab
343 when searching for filesystem paths. Also fixed some problems
344 caused by binding mounts or multiple block device mounts.
345 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
346 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
347 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
348 and #160305, Sourceforge bugs #618699 and #679832).
350 5. Made dump's -I option accept the value '0' meaning all the
351 read errors will be ignored. This can be useful when running
352 dump from unattended sessions (like cron jobs). Thanks to
353 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
355 6. Fixed the output of dump to indicate 'blocks' instead of
356 'tape blocks' in the various messages (blocks are always
357 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
358 made it clearly print the current blocksize at the start of
359 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
362 7. Made rmt understand long (largefiles) seeks.
364 8. Fixed build with very old versions of libext2fs, where
365 EXT2_FT_* constants were undefined.
367 9. Made dump accept the dumpdates path on the command line
368 (-D file option) instead of using only the hardcoded one.
369 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
372 10. Enabled rmt, LFS, readline, QFA options by default in
373 ./configure. Updated the configure process (new versions
374 of config.guess, config.sub etc).
376 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
377 =======================================================================
379 1. Changed dump to use fcntl(F_SETLK) style locking instead
380 of flock() when locking the dumpdates file. With the old
381 locking scheme, a local user having read rights on the
382 dumpdates file could be able to do a Denial of Service attack
383 on dump. In order to lock the dumpdates file with the new
384 scheme, the user would need to have write access on the file.
385 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
386 reporting the bug (originally a bugtraq post).
388 2. Fixed interactive 'ls' which caused spurious errors warnings
389 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
390 <jorgen_ostling@users.sourceforge.net> for reporting this
393 3. Fixed dump's estimate when dealing with sparse inodes.
395 4. Modified dump to allow setting a blocksize bigger than 32kB
396 (raised the limit to 1024kB), because newer hardware needs
397 this for proper operation. Thanks to Dirk Traenapp
398 <dtraenapp@users.sourceforge.net> for reporting this.
400 5. Fixed a bug causing Dump to stop and report an error if an
401 inode number in the exclude file was followed by some amount
402 of whitespace. Thanks to Jeffrey Sofferin
403 <sofferin@users.sourceforge.net> for reporting this bug.
405 6. Fixed a bug which caused restore, in some particular cases,
406 to ask some 'scary' questions and leave a bunch of RSTTMP
407 directories behind when restoring incremental tapes. Thanks
408 to Philippe Troin <phil@fifi.org> for reporting this bug and
409 providing the test cases.
411 7. Changed the wording when inodes are excluded from dump:
412 replaced 'Added inode 7 to exclude list' with
413 'Excluding inode 7 (resize inode) from dump', as suggested
414 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
416 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
417 ===================================================================
419 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
420 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
421 <eros@lamel.bo.cnr.it> for reporting the bug.
423 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
424 ===================================================================
426 1. Made dump print out the ext2 logical block number in case of a read
427 error, which can be given as an argument to debugfs' ncheck command,
428 in order to find out the affected inode. Added note about this
429 usage in dump's man page.
431 2. Fixed a problem in restore when reading tapes written on big
432 endian machines with very old versions of dump. The patch was
433 contributed by George Helffrich <george@geology.bristol.ac.uk>.
435 3. Fixed the tape length calculation when using large tapes
436 and compression. Thanks to Georg Lippold
437 <g_lippold@sourceforge.net> for reporting the bug.
439 4. Added a new set of examples in dump_on_cd_2 directory, based
440 on dump_on_cd examples but somewhat enhanced, supporting DVD
441 media, and localized in english and german. Thanks to
442 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
444 5. Made dump save 32 bit UID/GID. Internally, this was achieved
445 by switching from the old BSD inode format to the new BSD
446 inode format, which means that the tape format was changed.
447 However, since all restore versions out there should
448 transparently support both inode formats, the change should
449 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
452 6. Fixed a lot of warnings in the code shown when compiling
453 with 'gcc -W'. Thanks to Matthias Andree
454 <matthias.andree@stud.uni-dortmund.de> for reporting this.
456 7. Fixed a small markup bug in the dump man page. Thanks to
457 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
460 8. Rewrote entirely the man pages using the tmac.an macro
461 package (Linux man page format) instead of the original BSD
462 format. They should be now cleaner and easier to modify.
464 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
465 ==================================================================
467 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
468 non Linux clients were misinterpreted. The description of the problem
469 (incompatible numbering in Linux mtio opcodes) is documented at
470 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
471 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
472 providing an excellent, cross-platform replacement for rmt in his
475 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
476 which caused restore to fail when the inode maps were not entirely
477 contained on the first volume (when using really small volumes or
478 when dumping a huge number of inodes).
480 3. Fixed a problem in dump, where files in subdirectories of directories
481 that have the nodump flag set (or being excluded with -e/-E) were
482 sometimes incorrectly dumped (instead of being excluded). The original
483 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
484 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
485 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
486 help testing the patch.
488 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
489 which makes one able to pipe the output of dump, by the net, to
490 a remote CD-burner server.
492 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
493 and use GNU's symbolic syntax over rmt instead of numerical values
494 to assure multiple platform compatibility.
496 6. Documented the -d option in restore.
498 7. Added a -v (verbose) mode to dump. For now it just prints the number
499 of the inode being dumped, but this could evolve in future versions
500 to include interesting debugging output.
502 8. Added a -o flag to restore, which automatically restores the current
503 directory permissions in -i or -x mode, without asking the operator
504 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
505 and Peter Samuel <peters@e-smith.com>.
507 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
508 ====================================================================
510 1. Fixed a bug in the QFA feature which could have caused
511 a bad tape positionning on restore (causing extra delays in
512 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
514 2. Added a small note in the dump man page specifying that
515 there should be no white space between the option letter and
516 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
518 3. Made restore work with older versions of the readline library,
519 by checking in configure for several extended symbols. Restore
520 can now be compiled with a readline as old as the 2.0 release
521 (though it may be a good idea to upgrade it to a more recent
522 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
523 reporting the build failures.
525 4. Fixed a performance problem with the QFA file creation in
526 dump, which made unnecessary seeks on the tape slowing down
527 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
528 reporting this issue.
530 5. Removed the inclusion of some kernel headers in the dump
531 source, which prevented the compile in some kernel/glibc
532 headers/architecture combination. Thanks to Bdale Garbee
533 <bdale@gag.com> for reporting the bug.
535 6. Added the appropriate error message when dump fails to
536 open the output file for writing. Thanks to Amith Varghese
537 <amithv@yahoo.com> for reporting this bug.
539 7. Made restore able to understand large Solaris ufsdump tapes
540 (containing inodes bigger than 4194304). Sun have introduced
541 an "extension" to the dump tape format when dealing with
542 those inodes, which was uncorrectly handled by Linux restore.
543 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
544 providing a test case.
546 8. Added the -m parameter to dump which optimises the output for
547 inodes having been changed but not modified since the last dump
548 ('changed' and 'modified' have the meaning defined in stat(2)).
549 For those inodes, dump will save only the metadata, instead of
550 saving the entire inode contents. Inodes which are either
551 directories or have been modified since the last dump are saved
552 in a regular way. Uses of this flag must be consistent, meaning
553 that either every dump in an incremental dump set have the flag,
556 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
557 =======================================================================
559 1. Fixed behaviour of dump when exceeding resource limits
562 2. Added the -L flag to restore to allow the user to specify a
563 maximal allowed number of miscompares when using restore
564 with the -C option to check the backup.
566 3. Detailed the manual entry for the -N option of restore.
568 4. Added the -a flag to restore to make able doing unattended
569 restores in -i or -x mode (automatically walks through the
570 multiple dump volumes).
572 5. Extended the QFA mode to work with local files and/or
573 remote tapes and files. This way, restore can know in advance
574 the tape number and the offset for the inodes to extract and
575 can minimize the extraction time by seeking directly to the
578 6. Added the -A <archive> option to both dump and restore,
579 which makes dump to archive a dump table-of-contents in
580 the specified file to be used by restore to determine
581 whether a file is in the dump file that is being restored.
582 (the archive file syntax is also compatible with the
583 Solaris ufsdump generated one).
585 7. Small fix in restore making it able to read some (broken ?)
586 Solaris ufsdump tapes.
588 8. Fixed dump to correctly recognise the root filesystem when using
589 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
590 for reporting this bug.
592 9. Added the -P <file> option to restore to create a
593 Quick File Access file from an already made dump. Patch
594 contributed by Uwe Gohlke <uwe@ugsoft.de>.
596 10. Made restore compile and run on Solaris, making it a
597 possible replacement for the standard ufsrestore. Port was
598 contributed by Uwe Gohlke <uwe@ugsoft.de>.
600 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
601 =====================================================================
603 1. Added a set of backup scripts from Eugenio Diaz
604 <getnito@yahoo.com> in the examples section. It features
605 automatic (cron based) full and incremental dumping of
606 several filesystems on a separate filesystem.
608 2. Fixed a off-by-one miscalculation which disabled dumping
609 a one letter subdirectory. Thanks to Chris Siebenmann
610 <cks@utcc.utoronto.ca> for reporting the bug.
612 3. Fixed several restore bugs occuring when trying to
613 restore some missing files on the tape. Thanks to Chris
614 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
616 4. Fixed --with-ldopts configure argument passing, installing from
617 a separate object directory, makefile cleanups contributed
618 by <splite@purdue.edu>.
620 5. Fix a bug which could caused, in some conditions, the highest
621 number inode of a filesystem, to not be dumped. Many thanks
622 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
625 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
626 =======================================================================
628 1. Added a mini howto from Patrick Walsh in the examples
631 2. Minor man pages syntax corrections. Thanks to
632 Chris Danis <screechco@home.com> for reporting the bugs.
634 3. Added a script from David B. Peterson <dave@toppledwagon.com>
635 to the examples section. It features dumping several
636 filesystems to a remote tape drive upon ssh.
638 4. Added a patch provided by Richard Jones <rich@annexia.org>
639 which allows BRADEMAX (number of read errors tolerated by
640 dump) to be adjusted using the -I option.
642 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
643 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
646 6. Add the -l option to restore to specify if, when doing a
647 remote restore, the file used to access the backup is a
648 regular file (the defaults being a tape drive). Restore needs
649 to know this information when reading a remote compressed
650 dump. Previously, this information was autodetected, but
651 the autodetection code fails (with ioctl: Inappropriate ioctl
652 for device) when using a non Linux remote box. Thanks to
653 many users and especially to Eros Albertazzi
654 <eros@lamel.bo.cnr.it> for reporting this.
656 7. Found a workaroung for the dump deadlock problem (3 childs
657 stuck in pause(), father in read()). The workaround seems
658 to work for me and several beta-testers. If it doesn't work
659 for you, please report back.
661 8. Updated the RPM spec file (BuildPrereq, URL etc).
663 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
664 ========================================================================
666 1. Fixed the permissions of a newly created QFA file by dump.
668 2. Cleaned up the source of dump (the external variables
669 definition was a complete mess, making possible to have
672 3. Fixed restore to use the full tape volume path when doing
673 a compare (since it changes the working directory to the
674 filesystem being compared in the process).
676 4. Added the -q option to dump which makes dump abort
677 whenever it needs operator attention. It should be
678 easier to use dump in scripts with this option.
680 5. Detect the use of incompatible options to dump and
681 refuse them (like -a and -B options together).
683 6. Added bzip2 compression to dump/restore (use option -j level
684 to select it). Note that you will need a restore version
685 >= 0.4b24 in order to restore a bzip2 compressed dump.
686 The same warning as for the zlib compression applies:
687 the tape format of a bzip2 dump is not compatible with the
688 original BSD tape format.
690 7. Fixed a overflow problem in dump corrupting the dump when
691 very large files were encountered. Thanks to Vladimir Nadvornik
694 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
695 kernel buffer/page cache before starting the dump, helping
696 a bit those who use dump on mounted filesystems. Thanks to
697 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
700 9. Updated the RPM spec file following the RedHat changes
701 (dynamically linked binaries now in /usr/sbin etc).
703 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
704 which enables restore to recognise multi volume compressed dumps
705 done on CD. Included his scripts for dump (which pipe the dump
706 output directly into cdrecord) and restore. There is now
707 possible to to backups to CD on the fly!
709 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
710 ===================================================================
712 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
713 Marcus Meissner (Caldera International Security Dept.).
715 2. Implement the Sun rmt extended protocol. Patch submitted
716 by Ian Gordon <iangordon@users.sourceforge.net>.
718 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
719 the linux kernel header. This ensures that dump always has the
720 latest version of this file. Patch submitted by
721 Andreas Dilger <adilger@turbolinux.com>.
723 4. Report any filesystem present in either /etc/fstab with a
724 positive passno or /etc/dumpdates in dump -w output.
725 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
727 5. Fixed the looping problem in dump introduced in the
730 6. Changed the -B option of dump to limit the size of
731 _compressed_ output per volume if compression is on.
732 Patch contributed by Helmut Jarausch
733 <jarausch@igpm.rwth-aachen.de>. Note however that, since
734 it is impossible to predict the size of the compressed
735 data before doing the compression, the -B limit is a bit
738 7. Fixed a bug in reading the operator typed file/tape path for
739 the next volume in restore.
741 8. Implemented a "-F script" option to restore which permits the
742 user to specify a script which will be launched at the
743 beginning of each tape, useful for automatic programming of
744 tape changers for example. See the restore man page for the
745 script parameters and return codes.
747 9. Small fix for the QFA routines provided by Uwe Gohlke
748 <uwe@ugsoft.de>, and some recommendations for QFA uses in
751 10. Fixed the multivolume restoring where making a mistake
752 to the 'Mount next tape' prompt caused several blocks to
755 11. Enhanced the -e option of dump to take as a parameter a
756 comma separated list of inode numbers.
758 12. Added the -E option to dump which specify a file containing
759 inode numbers to exclude from the dump.
761 13. Fixed the compressed multi-volume dump + restore.
763 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
764 ==================================================================
766 1. Made dump report the number of blocks dumped per volume.
767 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
769 2. Fix a bug in dump where waiting too much at the 'change volumes'
770 question caused the volume to be rewritten. Thanks to
771 Graham Reed <greed@users.sourceforge.net> for reporting the
772 bug and providing a patch.
774 3. Added a compression option to dump, contributed by Jerry
775 Peters <gapeters@worldnet.att.net>.
777 WARNING: the compression works only on files/pipes and
778 drives supporting variable block size.
780 WARNING2: tapes created using the compression feature are
781 incompatible with the standard dump tape format, and a
782 version of restore >= 0.4b22 will be needed for extraction.
784 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
785 architectures. Thanks to Paul Slootman <paul@debian.org> for
786 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
789 5. Many cleanups (CPP defines, const char warnings, check of
790 ext2fs COMPAT flags, time_t cleanups, added libext2 version
791 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
793 6. Made --prefix option work in configure. All the install path
794 are now based on the configure parameters.
796 7. Added the Quick File Access mode in dump/restore, contributed
797 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
798 a file tape position for each inode, and this file is used by
799 restore (if called with parameter Q and the filename)
800 to directly position the tape at the file restore is currently
801 working on. This saves hours when restoring single files from
802 large backups, saves the tapes and the drive's head. Use
803 --enable-qfa option of configure to compile in the QFA support.
805 8. Added the possibility to dump several files and directories
806 in a single invocation of dump. Thanks to Uwe Gohlke
807 <uwe@ugsoft.de> for implementing this option.
809 9. Fixed the dumping and restoring of files with holes
810 (files ending with holes didn't get dumped or restored
813 10. Fixed a socket descriptor leak in dump, which leaved opened
814 3 file descriptors per dump process (and there is one dump
817 11. Fixed dump large file system support, by explicit use of
818 open64/lseek64/etc functions (explicit use needed because
819 e2fsprogs libraries don't behave well when compiled with
820 FILE_OFFSET_BITS=64).
822 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
823 ======================================================================
825 1. Fixed some bugs in the dump -w|-W logic introduced by
826 the previous version. Thanks to Andreas Dilger
827 <adilger@turbolinux.com> for his help on this one.
829 2. Fixed again a compilation problem when using old e2fs
830 headers (filesystem label related). Thanks to many users
831 who reported this stupid error.
833 3. Fixed a build problem on old lib5 based systems dealing with
834 _PATH_MNTTAB being undefined. Thanks to John Adams
835 <johna@onevista.com> for reporting the bug.
837 4. Improved the error detection in restore when creating
838 the temporary files in TMPDIR. Now restore will corectly
839 report a 'no space left on device' error instead of
840 strange errors which could imply an archive corruption.
841 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
842 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
844 5. Added the throughput information in dump's progression
845 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
848 6. Use libext2fs's inode scanning routines, which are particularly
849 robust when dealing with errors (especially when having some
850 bad blocks on the filesystem). Many thanks to Colin
851 <colin@horizon.com> for reporting the bug, investigating
852 the issues, finding a workaround, writing the patch and
853 fully testing it... (of course, if this patch will break
854 anything, he is to blame for :-)).
856 7. Made dump and restore LFS aware. Dump can dump a filesystem
857 containing large files, generate a large file on output and
858 restore can restore them. This feature is not enabled by
859 default, use --enable-largefile option of configure to enable
860 it (you need to have a LFS aware glibc though). Thanks to
861 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
862 and to Theodore T'so <tytso@valinux.com> for his always
865 8. Made dump ask upon a tape write error if it should rewrite
866 the current volume (assume this is a bad tape) or if it should
867 assume an end-of-tape condition (useful for tape drives which
868 behaves badly at the end of the tape). Thanks to Andreas
869 Dilger <adilger@turbolinux.com> for the suggestion.
871 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
872 =======================================================================
874 1. Fixed a small compilation problem due to a change
875 in the definintion of the struct sigaction in
876 glibc 2.0 / libc5. Thanks to Gunther Schlegel
877 <schlegel@riege.de> for reporting the bug and to
878 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
881 2. Modified the label and uuid analysis in order to be
882 self-contained (independant of kernel/libc headers). This
883 should fix the compile with older kernel/libc headers and
884 will preserve the functionality. Thanks to Bernhard Erdmann
885 <bernhard.erdmann@gmx.de> for reporting the bug.
887 3. The 'exclude inode' option, if applied to a directory
888 inode, excludes now correctly all the leaves of this
889 directory. Thanks to John R. Dennison
890 <gerdesas@users.sourceforge.net> for reporting the bug.
892 4. Fixed the '-e' option to disable the possibility
893 to exclude the root inode (which causes the dumps to
894 be unreadable by restore). Prevented array overflow
895 when multiple -e options are used.
897 5. Fixed dump to correctly interpret a filesystem argument
898 which is a mountpoint and it is not an absolute path
899 (as specified in the fstab). Thanks to Bernhard R. Erdmann
900 <be@berdmann.de> for reporting the bug.
902 6. Made dump able to backup files larger than 2 GB. Note that
903 dump still doesn't cope with files larger than 4 GB.
905 7. Restore the real uid and gid before invoking an external
906 RSH program (big hole when dump or restore is suid root!).
908 8. Get the values for the default tape device and the location
909 of fstab file from the system headers. Thanks to
910 Andreas Dilger <adilger@turbolinux.com> for the patch.
912 9. Made dump -w|-W report all recognized filesystems
913 present in either /etc/fstab or /etc/dumpdates, and present
914 the list in the same order as in fstab file. Thanks
915 to Andreas Dilger <adilger@turbolinux.com> for the patch.
917 10. Made dump's -a (automatic end of tape detection) the
918 default. Specifying one of -B, -c, -d or -s options will
919 override the EOM detection. Thanks to Andreas Dilger
920 <adilger@turbolinux.com> for the patch.
922 11. Save the ext2 filesystem label into the dump volume label.
923 Specifying a volume label on the command line (-L option)
924 will override this feature. Thanks to Andreas Dilger
925 <adilger@turbolinux.com> for the patch.
927 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
928 =====================================================================
930 1. Fixed the signal handling in dump (which I broke in 0.4b17)
931 which was causing several strange dump failures (dump
932 hanged or segmentation faults etc).
934 2. Specified the default blocksize in the dump man page.
936 3. Changed two info messages of restore to be written on stdout
937 instead of stderr in order to leave stderr only for errors
938 or warnings. Thanks to Stephen Atwell
939 <satwell@urbana.css.mot.com> for the suggestion.
941 4. Corrected an off by one calculation which prevented
942 dumping a directory having a 1 character name length.
943 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
944 for reporting the bug.
946 5. Reinforce dump in order to allow only level 0 and no
947 -u option when dumping a subdirectory, and document
948 this in the man page. Thanks to Bernhard Erdmann
949 <bernhard.erdmann@gmx.de> for reporting the bug.
951 6. Fixed a small harmless bug in restore which caused
952 in some conditions hard links to be restored several
953 times (therefore generation some warning messages).
954 Thanks to Kenneth Porter <shiva@well.com> for
957 7. Updated the RPM spec file to the latest RedHat version,
958 providing FHS packaging and other cosmetic changes.
959 You will need RPM version >= 3.0.5 in order to build the RPM.
961 8. Updated the configure script to check for libtermcap
962 before checking for libreadline, because we need this
963 library in order to compile the readline support.
965 9. Made dump understand the LABEL= and UUID= notation
966 both in /etc/fstab and on the command line. Note that
967 you will need the /proc filesystem in order to use
968 these notations. Thanks to Erik Troan <ewt@redhat.com>
969 for providing the patch.
971 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
972 ===================================================================
974 1. Fixed a potential buffer overflow in restore. Thanks
975 to Stan Bubrouski <satan@fastdial.net> for reporting
978 2. Fixed a readline-related bug which prevented
979 'cat DUMPFILE | restore -i -f -' from working. Thanks
980 to Charles Karney <karney@users.sourceforge.net>
983 3. Changed a few "panic" into "exit", causing restore to
984 be more stable against some attacks (like the last one
985 reported on Bugtraq, although the last version of restore
986 was not vulnerable - just dumped core). Thanks to
987 Andreas Hasenack <andreas@conectiva.com.br> for reporting
990 4. Removed the suid-root bit on dump and restore in the
991 default build (and generated RPMs). It should be safer
992 now. Users who need the suid-root capabilities in order
993 to make network backups should read first the man page
994 of dump and enable the suid bit manually.
996 5. Added -ltermcap to the compile parameters for restore
997 when using readline support, in order to make the compile
998 process work on some systems (SuSE ?). Thanks to
999 Patrik Schindler <poc@pocnet.net> for reporting the bug.
1001 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
1002 ====================================================================
1004 1. The -F script is called now *only* at the end of a tape,
1005 not at the end of the dump. Thanks to Christian Haul
1006 <haul@informatik.tu-darmstadt.de> for the bug report.
1008 Normally, the device name and the current volume number
1009 are passed on the command line of the script. However,
1010 if you want the old style script (with no arguments
1011 passed on the command line), you can enable it in
1012 configure with the --enable-oldstylefscript.
1014 2. Use posix signal handling to preserve dump functionality
1015 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
1018 3. Made the exit code of restore in case of a 'C'ompare
1019 command reflect the result of the compare. An exit status
1020 of 0 means the dump archive is correct, 1 means tape errors,
1021 2 means that some files were modified. Thanks to Kenneth Porter
1022 <shiva@well.com> for the suggestion.
1024 4. Made (finally) quotes work in interactive restore.
1026 5. Small fixes in order to allow dump to compile with a
1027 really old e2fsprogs version. Thanks to Ian Zimmerman
1028 <itz@speakeasy.org> for the bug report.
1030 6. Add GNU readline capabilities to interactive restore.
1031 Use configure's --enable-readline flag to enable this feature.
1032 Thanks to Patrik Schindler <poc@pocnet.net> for the
1035 7. Do the compare on the fly in restore 'C' mode (this will
1036 allow not to exhaust the available /tmp space when
1037 ccmparing large files). Thanks to Kenneth Porter
1038 <shiva@well.com> for the suggestion.
1040 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
1041 ====================================================================
1043 1. Fixed some several duplicate 'DUMP: DUMP:' in the
1046 2. Corrected the estimation of blocks to dump. Note that
1047 this is still wrong for large directory inodes, due
1048 to the size of a BSD directory entry on the tape
1049 (estimation is based on the size of the ext2 directory,
1050 which is smaller because it doesn't contain the null
1051 character at the end).
1053 3. Made dump report the total number of MB written to
1054 tape at the end of run. Thanks to W. Reilly Cooley
1055 <wcooley@nakedape.cc> for the patch.
1057 4. Added the -X option to restore giving the possibility
1058 to read the names of the files to be extracted/listed
1059 from a text file (in addition of the command line).
1060 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
1063 5. Added the device name and the current volume number
1064 as arguments to the end of tape script (dump -F option).
1066 6. Made the multi-volume dump work again (after having
1067 broken it in 0.4b15).
1069 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
1070 ===================================================================
1072 1. Added a prompt command in interactive restore mode. Thanks
1073 to Andreas Dilger <adilger@home.com> for the patch.
1075 2. Fixed a buffer overflow problem in dump (caused by
1076 not checking the size of the filesystem parameter).
1077 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
1078 reporting this on Bugtraq (and to several dump users
1079 who forwarded me his mail).
1081 3. Added the '-F script' option to dump in order to
1082 launch a script at the end of each tape (to be used
1083 with a tape changer, or to notify the sysadmin by
1086 4. Fixed a bug in restore compare code caused by the changes
1089 5. Fixed the treatment of options using the old BSD syntax
1090 in both dump and restore.
1092 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
1093 =======================================================================
1095 1. Fixed a bug in dump which may have caused invalid deleted
1096 directories to be dumped out if they were deleted after the
1097 mapping pass. This could occure on active filesystem and lead
1098 to heap corruption (causing dump malfunction in many possible ways).
1099 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
1100 bug and submitting the patch.
1102 2. Fixed the handling of the filesystem parameter in dump. A
1103 '/mnt/disk' parameter caused the disk contents to be dumped,
1104 but a '/mnt/disk/' parameter caused the mountpoint directory
1105 to be dumped (generally an empty directory).
1107 3. Improved the output of dump in order to tell which directory
1108 it is currently dumping (when dumping a subtree).
1110 4. Added the '-e' exclude inode option to dump. Thanks to
1111 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
1113 5. Added a REPORTING-BUGS file in order to provide a guide
1114 on how to correctly report a bug in dump/restore.
1116 6. Fixed a restore problem when restoring a hard link to an inode
1117 having the immutable or append-only attribute set. Thanks to
1118 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
1120 7. Fixed a compatibility problem between dumps made on little
1121 endian machines (the format was correct) and big endian
1122 machines (the format was incorrect). This fix break the
1123 compatibility with the older dumps made on big endian
1124 machines (sparc, m86k, ppc etc). For the first time in
1125 linux dump's history (I believe), the dumps made by *BSD,
1126 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
1127 so interchangeable. Thanks to Rob Cermak
1128 <cermak@ahab.rutgers.edu> for submitting the bug and
1129 helping me test the fix.
1131 8. Fixed the way dump reports the remaining percent/time, if
1132 the number of blocks actually dumped exceeds the estimated
1133 number of blocks. Thanks to Jean-Paul van der Jagt
1134 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
1136 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
1137 ======================================================================
1139 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
1140 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
1142 2. Removed the requirement to build the RPM as root from the
1143 spec file. Thanks to Christian Weisgerber
1144 <naddy@mips.rhein-neckar.de> for submitting this.
1146 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
1147 causing dump to block when dumping really huge filesystems.
1148 Many thanks to Patrik Schindler <poc@pocnet.net> for
1149 helping me find this bug.
1151 4. Fixed the treatment for an interrupt signal when dump access
1152 the remote tape through RSH. Thanks to Christian Weisgerber
1153 <naddy@mips.rhein-neckar.de> for providing the patch.
1155 5. Fixed a bug which was causing dump/restore to display
1156 garbage characters instead of the remote host name.
1158 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
1159 =====================================================================
1161 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
1162 <kukuk@suse.de> for submitting the patch.
1164 2. Fix for the exit code when using the size estimate option of
1165 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
1166 submitting the patch.
1168 3. Handle EINTR in atomical reads/writes in dump, which was causing
1169 dump to fail on some systems. Thanks to Eric Jergensen
1170 <eric@dvns.com> for reporting the bug and submitting the patch.
1172 4. Handle more than 16 characters for the device names in dumpdates.
1173 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
1174 tracking down the problem and proposing the solution.
1176 5. Fixed a bug in dump which prevented the creation of the
1177 dumpdates file when doing a 0-level dump without already
1178 having a dumpdates file. Thanks to Patrik Schindler
1179 <poc@pocnet.net> for reporting the bug.
1181 6. Changed the way dump 'S' flag reports the size estimate
1182 from number of blocks into bytes (making it compatible
1183 with the Solaris version, and simplifying things for
1184 amanda users). Thanks to Jason L Tibbitts III
1185 <tibbs@math.uh.edu> for reporting the bug.
1187 7. Fixed a compatibility problem in linux/alpha dump tape format.
1188 Now the linux/alpha dump are (again) compatible with the
1189 other dump formats. But this breaks compatibility with
1190 older dumps made on alpha. Thanks to Mike Tibor
1191 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
1193 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
1194 ======================================================================
1196 1. Added a '--enable-kerberos' to configure.
1198 2. Added a 'S' option to dump which determines the amount of space
1199 that is needed to perform the dump without actually doing it, similar
1200 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
1201 <cermak@ahab.rutgers.edu>.
1203 3. Added a 'M' multi-volume option to dump and restore which enables
1204 dumping to multiple files (useful when dumping to an ext2
1205 partition to make several dump files in order to bypass the 2GB
1206 file size limitation). The 'f' argument is treated as a prefix and
1207 the output files will be named <prefix>001, <prefix>002 etc. With
1208 the 'M' flag, restore automatically selects the right file without
1209 asking to enter a new tape each time.
1211 4. Fixed a memory leak which was causing dump to grow very big
1212 (270MB when dumping a 10GB filesystem...). Thanks to Jason
1213 Fearon <jasonf@netrider.org.au> for reporting the bug.
1215 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
1216 ======================================================================
1218 1. Make configure test if the system glob routines support
1219 extended syntax (ALTDIRFUNC). If not, use the internal glob
1220 routines instead of system ones. Thanks to Bernhard Sadlowski
1221 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
1222 and helping me resolve this and other minor libc5 compiling
1225 2. Fix a problem when dumping a ext2fs with the 'filetype'
1226 feature enabled. Thanks to Patrick J. LoPresti
1227 <patl@cag.lcs.mit.edu> for reporting the bug and to
1228 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
1230 3. Made the nodump flag work on directories. A directory which
1231 has the nodump flag gets never dumped, regardless of its
1234 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
1235 which allows dump on an active ext3 filesystem. However, this
1236 is a "quick and dirty" patch which enables backup of an ext3
1237 filesystem through the ext2 compatibility (by ignoring the
1238 NEEDS_RECOVERY bit). The journal file is not recognized and
1239 it is dumped (it should not).
1241 5. Test the superblock compatibility flags when dumping, in order
1242 to be sure that we know how to deal with specific features.
1244 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
1245 ====================================================================
1247 1. Use lchown instead of chown, fixing a possible security problem
1248 when restoring symlinks (a malicious user could use this
1249 to deliberately corrupt the ownership of important system files).
1250 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
1251 this and providing the patch.
1253 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
1254 ====================================================================
1256 1. Put dump sources under CVS, added Id tags in all files so
1257 one can use 'ident' on binary files.
1259 2. Added the dump/restore version in the usage text so one can
1260 easily verify the version he is using.
1262 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
1263 a va_start/va_end problem on linux-ppc (always call va_start
1264 va_end in pairs each time we use a vararg function).
1266 4. Added again the DT_* constants because old libc does not
1267 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
1268 for submitting the bug report.
1270 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
1271 this should enable dumping big (huge) filesystems.
1273 6. Added the RSH environment variable in order to be able to
1274 use a rsh replacement like ssh when doing remote backups (and
1275 bypass the security limitations of rcmd). Now you can do remote
1276 backups without being root (or making dump setuid root).
1278 7. Modified again the way dumpdates works. For incremental dumps,
1279 we need to read dumpdates even if we are not using 'u' option.
1280 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
1283 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
1284 ===================================================================
1286 1. Removed the 'k' flag from the restore 'about' text if kerberos
1287 was not compiled in.
1289 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
1290 to them (fsetflags takes a char*, setflags an open fd!).
1292 3. (f)setflags is called only if the flags aren't empty. If the
1293 file is a special file, a warning is printed, because changing
1294 flags implies opening the device. Normally, a special file
1295 should not have any flag... (Debian bug #29775, patch provided
1296 by Abhijit Dasgupta <abhijit@ans.net>).
1298 4. Made possible to dump a file system not mentioned in /etc/fstab.
1299 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
1301 5. Changed the default behaviour to not create dumpdates
1302 unless 'u' option is specified. Removed the old "debian-patch"
1303 which provided the same thing. (Debian bug #38136, #33818).
1305 6. Removed all those dump*announce, since they were getting old...
1307 7. Added warning messages if dumpdates does not exist and
1308 when an inferior level dump does not exist (except for a level 0
1311 8. Debugged the glob calls in interactive mode: restore used a
1312 dirent struct which was different from the /usr/include/dirent.h
1313 one (this used to work, is it a glibc2 change?), so none of the
1314 compat glob (which used /usr/include/dirent.h) or the system glob
1315 worked. Restore use now the system dirent (and the system
1316 DT_* constants), which are compatible with BSD ones.
1318 9. Added a configure flag (--with-dumpdatespath) to specify
1319 the location of dumpdates file. By default, it is
1322 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1323 included the current date/version in man pages.
1325 11. Corrected the estimation of remaining time when
1326 the operator doesn't change the tapes quickly enough. This
1327 was an old bug, I thought I corrected it, and discovered
1328 that in fact it was corrected in two different places, so
1329 the results canceled each other...
1331 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1332 ===================================================================
1334 1. Integrated multiple patches from RedHat, Debian and SuSE:
1336 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1338 - several fixes in the man pages.
1339 - update the default tape device to /dev/st0.
1340 - many updates for Linux Alpha (byte ordering, size_t etc).
1342 - use environment variable for TMPDIR (instead of /tmp).
1343 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1344 - workaround egcs bug (RedHat bugs #4281 and #2989).
1345 - wire $(OPT) throughout Makefile's.
1347 2. Upgrade the dump revision to 1, making possible to dump filesystems
1348 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1350 3. Fix some compile warnings, prototype all functions.
1352 4. Use glibc err/glob instead of internal compatibility
1353 routines (only if available).
1355 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1356 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1358 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1359 ======================================================================
1361 1. Integrated the changes from FreeBSD-3.1-RELEASE
1362 (mostly bug fixes, buffer overruns, dump has now an "automatic
1363 tape length calculation" flag, dump/restore can use kerberos now
1364 (this is NOT tested), use environment variables for TAPE and
1367 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1369 3. Corrected a bug in restore when using 'C' option with multi-volumes
1370 tapes (files splited accros two tapes give "size changed" errors
1373 4. Corrected the long standing bug when dumping multiple tapes.
1374 This works for me, needs further testing.
1376 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1377 ====================================================================
1379 1. Dump now runs correctly on kernels 2.1.x
1380 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1382 Changes between versions 0.4b2 and 0.4b3
1383 ========================================
1385 1. Use realpath() if available
1387 2. Report statistics
1389 Changes between versions 0.4b1 and 0.4b2
1390 ========================================
1392 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1395 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1397 3. Fixed configure.in to correctly handle --enable-rmt
1399 Changes between versions 0.3 and 0.4b1
1400 ======================================
1402 1. Integrated the changes from 4.4BSD-Lite2
1404 2. Integrated the patches from Debian and Red Hat
1406 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1408 4. Changed dump to use the Ext2fs library to get block addresses. This
1409 should solve the endianness problem on SparcLinux.
1411 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1412 distribution's one) to use autoconf
1414 6. Fixed a few minor bugs
1416 Changes between versions 0.2e and 0.2f
1417 ======================================
1419 1. Added the creation of named pipes (fifos) in restore.
1421 2. Added the -N flag in the restore manual page.
1423 3. Added the file kernel.patch which contains the llseek() optimization
1424 patch for 1.2.x kernels.
1426 4. Fixed a bug in the restoration of symbolic links: owner and group were
1429 5. Integrated some changes from FreeBSD 2.2.
1431 6. Added a call to ftruncate() after restoring each file to restore
1432 correctly files ending by a hole.
1434 Changes between versions 0.2d and 0.2e
1435 ======================================
1437 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1438 with owner = root (0) and group = root/wheel/whatever (0).
1440 Changes between versions 0.2c and 0.2d
1441 ======================================
1443 1. Dump is now able to backup 2GB+ filesystems.
1445 2. Dump and restore can now be linked as static binaries.
1447 Changes between versions 0.2b and 0.2c
1448 ======================================
1450 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1452 Changes between versions 0.2a and 0.2b
1453 ======================================
1455 1. Really fixed the bug that I should have corrected in 0.2a.
1457 2. Enabled optimization again.
1459 Changes between versions 0.2 and 0.2a
1460 =====================================
1462 1. Disabled the optimization during compilation.
1464 Changes between versions 0.1 and 0.2
1465 ====================================
1467 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1468 the fs_type field (actually, I modified the file fstab.c to make it
1469 use the mntent functions).
1471 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1472 made by dump should be readable by the BSD restore and backups made
1473 by the BSD dump should be readable by restore. Unfortunately, this
1474 means that the dump format has changed between version 0.1 and version
1477 3. Dump is now able to backup a subtree, it is no longer limited to whole
1478 filesystems like the BSD version.
1480 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1483 Changes between versions 0.0 and 0.1
1484 ====================================
1486 1. Now create links rdump and rrestore during the `make install' step.
1488 2. Linux port specific bugs added to the manual pages
1490 3. Incorrect estimation of the number of tapes blocks fixed when doing
1491 incremental backups.
1493 4. Better ls-like format in restore in interactive mode.
git.wh0rd.org / dump.git / blob