[dump.git] / CHANGES
1 $Id: CHANGES,v 1.269 2005/01/21 09:27:26 stelian Exp $
3 Changes between versions 0.4b38 and 0.4b39 (released January 21, 2005)
4 ======================================================================
6 1. The newly added dump_on_cd_3 example was buggy, replace it
7 with an updated version from Andrew Basterfield
8 <bob@cemetery.homeunix.org>.
10 2. Made restore to chdir() back into the initial directory when
11 dumping core while aborting a comparision operation. The
12 previous behaviour was to write the corefile at the root of
13 the directory being compared, which could very well be
14 read only and preventing the corefile generation. Thanks
15 to Kenneth Porter <shiva@sewingwitch.com> for the bug report.
17 3. Silenced the failure to call fgetflags() when comparing an
18 entry which has no ext2 attributes (as in lsattr()).
20 4. Fix a brown paper bug in restore -C which broke restore and
21 caused modifications on the filesystem being compared
22 (directories containing a file with the same name as the
23 directory get renamed to RSTTMP...). Thanks to Kenneth Porter
24 <shiva@sewingwitch.com> for finding the bug and helping me
25 reproduce it.
27 5. Made restore -C force the -N flag (no writing allowed on
28 the disk) in order to prevent more bugs like the above one.
30 Changes between versions 0.4b37 and 0.4b38 (released January 7, 2005)
31 =====================================================================
33 1. Fix a couple of troff syntax bugs in the man pages.
34 Thanks to Eric Raymond <esr@thyrsus.com> for the patch.
36 2. Made restore use either libncurses or libtermcap, depending
37 on which one is available at configure time.
39 3. Fixed restore negative size display bug when comparing a
40 dump containing files over 2GB. Thanks to Steve Bonds
41 <sbonds@users.sourceforge.net> for the bug report.
43 4. Do not save directory entries to non-dumped inodes
44 (excluded from dump). This will eliminate the 'missing
45 file' warnings when doing 'restore -C'.
47 5. Fix dump crash when backuping a huge (2TB) filesystem,
48 due to a bogus calculation on the inode map size.
49 Thanks to Kevin B. Haines <K.B.Haines@rl.ac.uk> for
50 submitting the bug and testing the fix.
52 6. Fix a problem in restore where the final \0 in the symbolic
53 link names could have been lost, generating corrupt filenames.
54 Thanks to Kyle Wilson <kyle.wilson@amd.com> for reporting the
55 bug.
57 7. Implemented a hash list for the directory names in restore.
58 The linear list used before caused problems in interactive
59 restores when dealing with directories having thousands of
60 entries. Thanks to Brian Ristuccia <bristuccia@starentnetworks.com>
61 for reporting the bug.
63 8. Improved restore -C, this time including the directory
64 attributes into the comparision.
66 9. Made restore understand tapes containing EA/ACLs (which will
67 be dumped by the next version of dump). In this version
68 extended attributes on the tape are ignored, for full EA/ACL
69 support wait for the next version or try the experimental EA
70 patch.
72 Changes between versions 0.4b36 and 0.4b37 (released July 7, 2004)
73 ==================================================================
75 1. Added the --enable-staticz configure option which enables
76 dump and restore to be linked with static versions of
77 libz and libbz2 (and dynamic versions of all other libraries).
78 This will make Debian users happy, because libz and libbz2
79 were the only needed libraries living in /usr, all the
80 others live in /lib. In case of system emergency, it is
81 better not to have to rely on an extra filesystem.
82 Thanks to Bdale Garbee <bdale@gag.com> for the suggestion.
84 2. Fix compilation on (at least the Linux Debian port to) AMD64.
85 (<ext2fs/ext2_types.h> defines some types (__s64 and __u64)
86 that are also defined by <linux/types.h> (<asm/types.h>) and
87 they conflict).
89 3. Make dump's reading of the dumpdates file a bit more robust,
90 preventing dump from crashing when the dumpdates file has
91 been modified by hand.
93 4. Fixed some offset calculations in dump code which could
94 lead to "bread lseek errors" on large filesystems. Thanks
95 to Bruce Lowekamp <lowekamp@users.sourceforge.net> for
96 reporting this bug and debugging the issue.
98 5. Made dump use the blkid library when searching for devices
99 by label or uuid instead of dump's own routines.
101 6. Corrected a bug in dump where a wrong LABEL=... line in
102 /etc/fstab could prevent dump from dumping unrelated
103 filesystems. Thanks to Bruce Lowekamp
104 <lowekamp@users.sourceforge.net> for reporting the bug.
106 Changes between versions 0.4b35 and 0.4b36 (released April 21, 2004)
107 ====================================================================
109 1. Fixed dump compilation with old gcc versions. Thanks to
110 Mike Castle <dalgoda@users.sourceforge.net> for the patch.
112 2. Fixed some warnings (howmany, roundup, powerof2 redefined)
113 when compiling against a recent glibc version.
115 3. Fixed a bug in restore preventing the read of a dump tape
116 written with Solaris 7 ufsdump. Thanks to Patrick Higgins
117 <phiggins@transzap.com> for reporting the bug and providing
118 the test case.
120 4. Changed dump to enable the creation of volumes of different
121 sizes in a single run (make -B accept a list of values).
122 Patch contributed by Florian Zumbiehl <florz@gmx.de>.
124 5. Use the glibc provided minor() and major() macros instead
125 of our own bitmask implementation. This should be safe for
126 when the major/minor namespace will migrate to 32 bits.
127 Thanks to Zhang Jun <zhangjun@nanjing-fnst.com> for reporting
128 the bug.
130 6. Made explicit in the dump man page that dump will not create
131 a remote file, it will only write to an already existing one.
133 7. Another try at making size estimates better again.
135 8. Put back the inconditional running of the end-of-tape
136 script which was changed in 0.434 to be run only when -M
137 or multiple -f were NOT specified. Some users rely on this
138 feature even when it is combined with -M/-f.
140 9. Fixed restore when restoring huge backups (where rstdir...
141 temporary files are over 2GB). Thanks to
142 Raphael Jaffey <rjaffey@artic.edu> for reporting this,
143 debugging the issue and making the patch.
145 10. Made restore understand FreeBSD UFS2 tapes. Thanks to
146 David <vrtin@users.sourceforge.net> for submitting the bug
147 and providing a test case.
149 11. Made dump work with an arbitrary integer level (no more 10
150 levels only limitation). Thanks to Matthew
151 <msvincen@midway.uchicago.edu> for the patch.
153 Changes between versions 0.4b34 and 0.4b35 (released December 21, 2003)
154 =======================================================================
156 1. Added a note in the dump man page saying that the default
157 blocksize can be 32 if -d is used with a high density argument.
158 Thanks to Antonios Christofides <A.Christofides@itia.ntua.gr>
159 for the patch.
161 2. Fixed configure to correctly understand CPPFLAGS, CFLAGS,
162 LDFLAGS environment variables. Thanks to Arcady Genkin
163 <antipode@users.sourceforge.net> for reporting the bug.
165 3. Made -e/-E options of dump accept an unlimited number of inodes
166 to be excluded, rather than a hardcoded maximum. Thanks to
167 Dietrich Rothe <d-rothe@users.sourceforge.net> for the patch.
169 4. Updated the autoconf system to 2.50. Forced the -D_BSD_SOURCE
170 and -D_USE_BSD_SIGNAL defines in configure in order to solve
171 64bit build problems because quad_t is redifined with a
172 different signature. Thanks to Mike Harris <mharris@redhat.com>
173 for reporting this bug.
175 5. Made restore build on Solaris, making possible to
176 restore Linux's "enhanced" tapes. Thanks to Uwe Gohlke
177 <uwe@ugsoft.de> for the patch.
179 6. Made an extension in the dump tape format capable of saving
180 MacOSX specific inode extensions. Uwe Gohlke <uwe@ugsoft.de>
181 wrote the extension and contributed the restore code back
182 into this codebase. The same extension mechanism will be
183 used in the future to save ACLs...
185 7. Made rmt work correctly with regard to QFA and local/remote
186 files and tapes. The remote access will however work only
187 when the dump provided rmt version is used. If you want to
188 use another rmt server, please do not use the QFA feature.
189 Thanks to Uwe Gohlke <uwe@ugsoft.de> for the patch.
191 Changes between versions 0.4b33 and 0.4b34 (released April 18, 2003)
192 ====================================================================
194 1. Fixed the INSTALL file to reflect the actual install paths.
195 Thanks to David Ronis <ronis@ronispc.chem.mcgill.ca> for
196 reporting the bug.
198 2. Fixed the configure script to only check for headers presence
199 instead of trying to compile them. This should fix issues
200 with old build environments. Thanks to Kari Lempiainen
201 <kari@funky.pp.fi> for reporting the bug.
203 3. Fixed restore to correctly ignore sockets when comparing
204 a dump (as socket cannot be properly restored anyway). Thanks
205 to Gunther Reiszig <gunther@mit.edu> for reporting the bug.
207 4. Fixed restore to correctly access the archive file (-A argument)
208 even when using a remote tape. Thanks to Fabrice Bellet
209 <fabrice@bellet.info> for reporting the bug.
211 5. Fixed (again) handling of long (largefile) seeks in rmt.
212 Thanks to Fabrice Bellet <fabrice@bellet.info> for reporting
213 the bug.
215 6. Fixed restore corner case when dealing with large block sizes
216 dump is able to write now (-b 1024). Thanks to Fabrice Bellet
217 <fabrice@bellet.info> for reporting the bug.
219 7. Fixed a bug preventing dump to access a filesystem having
220 a label exactly 16 bytes in length. Thanks to <minduni@ti-edu.ch>
221 for reporting the bug.
223 8. Made dump store dump dates using explicit timezones, fixing a
224 problem with incremental dumps if the timezone is modified
225 between the dumps. Thanks to Philippe Troin <phil@fifi.org> for
226 the bug report and the patch.
228 9. Fixed a bug encountered when dumping individual files (not full
229 filesystems or directories) and dangling symbolic links happen
230 to be in the list of files. For as far as dump is concerned,
231 dangling symbolic links are allowed, and are dumped as is.
232 Thanks to Jin-su Ahn <jsahn@ee.snu.ac.kr> for reporting the
233 bug and providing the fix.
235 10. Fixed open and creation modes and permissions for QFA and
236 table-of-contents files in dump and restore. Thanks to
237 Philippe Troin <phil@fifi.org> for the patch.
239 11. Fixed the archive file descriptor handling enabling it to be 0.
240 This can happen in some cases when shell redirections are used.
241 Thanks to Philippe Troin <phil@fifi.org> for the patch.
243 12. Delayed the opening of archive file until after suid had been
244 dropped (fixing a possible security issue if dump is suid).
245 Thanks to Philippe Troin <phil@fifi.org> for the patch.
247 13. Fixed the 'S' command handling in the rmt client part.
248 Thanks to Philippe Troin <phil@fifi.org> for the patch.
250 14. Modified the end-of-tape script handling to print out statistics
251 (and stop the timer) before launching the eot script. Also, the eot
252 script does not get run anymore when using -M (which makes sense) or
253 when multiple tapes are listed on the command line
254 (-f tape0,tape1,tapen) (which also makes sense).
255 Thanks to Philippe Troin <phil@fifi.org> for the patch.
257 15. Relicensed dump/restore under the 'revised' BSD license, as per
258 ftp://ftp.cs.berkeley.edu/ucb/4bsd/README.Impt.License.Change.
260 16. Added LZO compression to dump. This new compression method has
261 the advantage of being super fast, thus not killing tape streaming
262 on slow machines. Thanks to Helmut Jarausch
263 <jarausch@igpm.rwth-aachen.de> for the patch and to
264 Markus Oberhumer <markus@oberhumer.com> for giving special permission
265 to include his miniLZO project (GPL licensed) in dump/restore.
267 17. Some small buffer overruns fixes in rmt. Thanks to Antonomasia
268 <ant@notatla.demon.co.uk> for reporting the bugs.
270 18. Added a special rmt version which can do encryption when writing
271 to tape. Read examples/encrypted_rmt/README for details on
272 how to enable and configure it. Thanks to Ken Lalonde
273 <ken@globalremit.com> for the patch.
275 19. Made dump work with 2.5 kernel end of tape early warning semantics.
276 Thanks to Kai Makisara <Kai.Makisara@kolumbus.fi> for the patch.
278 20. Fixed a bug which caused dump -w|-W not to work anymore, because
279 the fs_freq and fs_passno fields in /etc/mtab are always set
280 to 0 0. Thanks to Trent Piepho <xyzzy@speakeasy.org> for
281 reporting the bug.
283 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
284 =======================================================================
286 1. Added a note in the restore man page clarifying the question
287 'set the permissions on the current directory ?' asked by
288 restore at the end of treatment in -i and -x modes.
290 2. Fixed the endianess issues when reading compressed tapes.
291 Thanks to Dark Force <daq4th@users.sourceforge.net> for
292 reporting this bug and providing test cases.
294 3. Fixed the "ACL won't be dumped" warning message (which showed
295 an extra, unrelated error message). Thanks to Dragan Krnic
296 <dkrnic@lycos.com> for reporting this bug.
298 4. Made dump look first into /etc/mtab, then into /etc/fstab
299 when searching for filesystem paths. Also fixed some problems
300 caused by binding mounts or multiple block device mounts.
301 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
302 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
303 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
304 and #160305, Sourceforge bugs #618699 and #679832).
306 5. Made dump's -I option accept the value '0' meaning all the
307 read errors will be ignored. This can be useful when running
308 dump from unattended sessions (like cron jobs). Thanks to
309 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
311 6. Fixed the output of dump to indicate 'blocks' instead of
312 'tape blocks' in the various messages (blocks are always
313 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
314 made it clearly print the current blocksize at the start of
315 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
316 the suggestions.
318 7. Made rmt understand long (largefiles) seeks.
320 8. Fixed build with very old versions of libext2fs, where
321 EXT2_FT_* constants were undefined.
323 9. Made dump accept the dumpdates path on the command line
324 (-D file option) instead of using only the hardcoded one.
325 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
326 suggestion.
328 10. Enabled rmt, LFS, readline, QFA options by default in
329 ./configure. Updated the configure process (new versions
330 of config.guess, config.sub etc).
332 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
333 =======================================================================
335 1. Changed dump to use fcntl(F_SETLK) style locking instead
336 of flock() when locking the dumpdates file. With the old
337 locking scheme, a local user having read rights on the
338 dumpdates file could be able to do a Denial of Service attack
339 on dump. In order to lock the dumpdates file with the new
340 scheme, the user would need to have write access on the file.
341 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
342 reporting the bug (originally a bugtraq post).
344 2. Fixed interactive 'ls' which caused spurious errors warnings
345 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
346 <jorgen_ostling@users.sourceforge.net> for reporting this
347 bug.
349 3. Fixed dump's estimate when dealing with sparse inodes.
351 4. Modified dump to allow setting a blocksize bigger than 32kB
352 (raised the limit to 1024kB), because newer hardware needs
353 this for proper operation. Thanks to Dirk Traenapp
354 <dtraenapp@users.sourceforge.net> for reporting this.
356 5. Fixed a bug causing Dump to stop and report an error if an
357 inode number in the exclude file was followed by some amount
358 of whitespace. Thanks to Jeffrey Sofferin
359 <sofferin@users.sourceforge.net> for reporting this bug.
361 6. Fixed a bug which caused restore, in some particular cases,
362 to ask some 'scary' questions and leave a bunch of RSTTMP
363 directories behind when restoring incremental tapes. Thanks
364 to Philippe Troin <phil@fifi.org> for reporting this bug and
365 providing the test cases.
367 7. Changed the wording when inodes are excluded from dump:
368 replaced 'Added inode 7 to exclude list' with
369 'Excluding inode 7 (resize inode) from dump', as suggested
370 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
372 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
373 ===================================================================
375 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
376 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
377 <eros@lamel.bo.cnr.it> for reporting the bug.
379 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
380 ===================================================================
382 1. Made dump print out the ext2 logical block number in case of a read
383 error, which can be given as an argument to debugfs' ncheck command,
384 in order to find out the affected inode. Added note about this
385 usage in dump's man page.
387 2. Fixed a problem in restore when reading tapes written on big
388 endian machines with very old versions of dump. The patch was
389 contributed by George Helffrich <george@geology.bristol.ac.uk>.
391 3. Fixed the tape length calculation when using large tapes
392 and compression. Thanks to Georg Lippold
393 <g_lippold@sourceforge.net> for reporting the bug.
395 4. Added a new set of examples in dump_on_cd_2 directory, based
396 on dump_on_cd examples but somewhat enhanced, supporting DVD
397 media, and localized in english and german. Thanks to
398 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
400 5. Made dump save 32 bit UID/GID. Internally, this was achieved
401 by switching from the old BSD inode format to the new BSD
402 inode format, which means that the tape format was changed.
403 However, since all restore versions out there should
404 transparently support both inode formats, the change should
405 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
406 reporting the bug.
408 6. Fixed a lot of warnings in the code shown when compiling
409 with 'gcc -W'. Thanks to Matthias Andree
410 <matthias.andree@stud.uni-dortmund.de> for reporting this.
412 7. Fixed a small markup bug in the dump man page. Thanks to
413 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
414 patch.
416 8. Rewrote entirely the man pages using the tmac.an macro
417 package (Linux man page format) instead of the original BSD
418 format. They should be now cleaner and easier to modify.
420 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
421 ==================================================================
423 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
424 non Linux clients were misinterpreted. The description of the problem
425 (incompatible numbering in Linux mtio opcodes) is documented at
426 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
427 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
428 providing an excellent, cross-platform replacement for rmt in his
429 star package.
431 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
432 which caused restore to fail when the inode maps were not entirely
433 contained on the first volume (when using really small volumes or
434 when dumping a huge number of inodes).
436 3. Fixed a problem in dump, where files in subdirectories of directories
437 that have the nodump flag set (or being excluded with -e/-E) were
438 sometimes incorrectly dumped (instead of being excluded). The original
439 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
440 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
441 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
442 help testing the patch.
444 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
445 which makes one able to pipe the output of dump, by the net, to
446 a remote CD-burner server.
448 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
449 and use GNU's symbolic syntax over rmt instead of numerical values
450 to assure multiple platform compatibility.
452 6. Documented the -d option in restore.
454 7. Added a -v (verbose) mode to dump. For now it just prints the number
455 of the inode being dumped, but this could evolve in future versions
456 to include interesting debugging output.
458 8. Added a -o flag to restore, which automatically restores the current
459 directory permissions in -i or -x mode, without asking the operator
460 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
461 and Peter Samuel <peters@e-smith.com>.
463 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
464 ====================================================================
466 1. Fixed a bug in the QFA feature which could have caused
467 a bad tape positionning on restore (causing extra delays in
468 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
470 2. Added a small note in the dump man page specifying that
471 there should be no white space between the option letter and
472 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
474 3. Made restore work with older versions of the readline library,
475 by checking in configure for several extended symbols. Restore
476 can now be compiled with a readline as old as the 2.0 release
477 (though it may be a good idea to upgrade it to a more recent
478 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
479 reporting the build failures.
481 4. Fixed a performance problem with the QFA file creation in
482 dump, which made unnecessary seeks on the tape slowing down
483 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
484 reporting this issue.
486 5. Removed the inclusion of some kernel headers in the dump
487 source, which prevented the compile in some kernel/glibc
488 headers/architecture combination. Thanks to Bdale Garbee
489 <bdale@gag.com> for reporting the bug.
491 6. Added the appropriate error message when dump fails to
492 open the output file for writing. Thanks to Amith Varghese
493 <amithv@yahoo.com> for reporting this bug.
495 7. Made restore able to understand large Solaris ufsdump tapes
496 (containing inodes bigger than 4194304). Sun have introduced
497 an "extension" to the dump tape format when dealing with
498 those inodes, which was uncorrectly handled by Linux restore.
499 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
500 providing a test case.
502 8. Added the -m parameter to dump which optimises the output for
503 inodes having been changed but not modified since the last dump
504 ('changed' and 'modified' have the meaning defined in stat(2)).
505 For those inodes, dump will save only the metadata, instead of
506 saving the entire inode contents. Inodes which are either
507 directories or have been modified since the last dump are saved
508 in a regular way. Uses of this flag must be consistent, meaning
509 that either every dump in an incremental dump set have the flag,
510 or no one has it.
512 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
513 =======================================================================
515 1. Fixed behaviour of dump when exceeding resource limits
516 (SIGXFSZ treatment).
518 2. Added the -L flag to restore to allow the user to specify a
519 maximal allowed number of miscompares when using restore
520 with the -C option to check the backup.
522 3. Detailed the manual entry for the -N option of restore.
524 4. Added the -a flag to restore to make able doing unattended
525 restores in -i or -x mode (automatically walks through the
526 multiple dump volumes).
528 5. Extended the QFA mode to work with local files and/or
529 remote tapes and files. This way, restore can know in advance
530 the tape number and the offset for the inodes to extract and
531 can minimize the extraction time by seeking directly to the
532 good tape/offset.
534 6. Added the -A <archive> option to both dump and restore,
535 which makes dump to archive a dump table-of-contents in
536 the specified file to be used by restore to determine
537 whether a file is in the dump file that is being restored.
538 (the archive file syntax is also compatible with the
539 Solaris ufsdump generated one).
541 7. Small fix in restore making it able to read some (broken ?)
542 Solaris ufsdump tapes.
544 8. Fixed dump to correctly recognise the root filesystem when using
545 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
546 for reporting this bug.
548 9. Added the -P <file> option to restore to create a
549 Quick File Access file from an already made dump. Patch
550 contributed by Uwe Gohlke <uwe@ugsoft.de>.
552 10. Made restore compile and run on Solaris, making it a
553 possible replacement for the standard ufsrestore. Port was
554 contributed by Uwe Gohlke <uwe@ugsoft.de>.
556 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
557 =====================================================================
559 1. Added a set of backup scripts from Eugenio Diaz
560 <getnito@yahoo.com> in the examples section. It features
561 automatic (cron based) full and incremental dumping of
562 several filesystems on a separate filesystem.
564 2. Fixed a off-by-one miscalculation which disabled dumping
565 a one letter subdirectory. Thanks to Chris Siebenmann
566 <cks@utcc.utoronto.ca> for reporting the bug.
568 3. Fixed several restore bugs occuring when trying to
569 restore some missing files on the tape. Thanks to Chris
570 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
572 4. Fixed --with-ldopts configure argument passing, installing from
573 a separate object directory, makefile cleanups contributed
574 by <splite@purdue.edu>.
576 5. Fix a bug which could caused, in some conditions, the highest
577 number inode of a filesystem, to not be dumped. Many thanks
578 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
579 chase this bug.
581 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
582 =======================================================================
584 1. Added a mini howto from Patrick Walsh in the examples
585 directory.
587 2. Minor man pages syntax corrections. Thanks to
588 Chris Danis <screechco@home.com> for reporting the bugs.
590 3. Added a script from David B. Peterson <dave@toppledwagon.com>
591 to the examples section. It features dumping several
592 filesystems to a remote tape drive upon ssh.
594 4. Added a patch provided by Richard Jones <rich@annexia.org>
595 which allows BRADEMAX (number of read errors tolerated by
596 dump) to be adjusted using the -I option.
598 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
599 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
600 patch.
602 6. Add the -l option to restore to specify if, when doing a
603 remote restore, the file used to access the backup is a
604 regular file (the defaults being a tape drive). Restore needs
605 to know this information when reading a remote compressed
606 dump. Previously, this information was autodetected, but
607 the autodetection code fails (with ioctl: Inappropriate ioctl
608 for device) when using a non Linux remote box. Thanks to
609 many users and especially to Eros Albertazzi
610 <eros@lamel.bo.cnr.it> for reporting this.
612 7. Found a workaroung for the dump deadlock problem (3 childs
613 stuck in pause(), father in read()). The workaround seems
614 to work for me and several beta-testers. If it doesn't work
615 for you, please report back.
617 8. Updated the RPM spec file (BuildPrereq, URL etc).
619 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
620 ========================================================================
622 1. Fixed the permissions of a newly created QFA file by dump.
624 2. Cleaned up the source of dump (the external variables
625 definition was a complete mess, making possible to have
626 objects overlap).
628 3. Fixed restore to use the full tape volume path when doing
629 a compare (since it changes the working directory to the
630 filesystem being compared in the process).
632 4. Added the -q option to dump which makes dump abort
633 whenever it needs operator attention. It should be
634 easier to use dump in scripts with this option.
636 5. Detect the use of incompatible options to dump and
637 refuse them (like -a and -B options together).
639 6. Added bzip2 compression to dump/restore (use option -j level
640 to select it). Note that you will need a restore version
641 >= 0.4b24 in order to restore a bzip2 compressed dump.
642 The same warning as for the zlib compression applies:
643 the tape format of a bzip2 dump is not compatible with the
644 original BSD tape format.
646 7. Fixed a overflow problem in dump corrupting the dump when
647 very large files were encountered. Thanks to Vladimir Nadvornik
648 for the bug report.
650 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
651 kernel buffer/page cache before starting the dump, helping
652 a bit those who use dump on mounted filesystems. Thanks to
653 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
654 for this suggestion.
656 9. Updated the RPM spec file following the RedHat changes
657 (dynamically linked binaries now in /usr/sbin etc).
659 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
660 which enables restore to recognise multi volume compressed dumps
661 done on CD. Included his scripts for dump (which pipe the dump
662 output directly into cdrecord) and restore. There is now
663 possible to to backups to CD on the fly!
665 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
666 ===================================================================
668 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
669 Marcus Meissner (Caldera International Security Dept.).
671 2. Implement the Sun rmt extended protocol. Patch submitted
672 by Ian Gordon <iangordon@users.sourceforge.net>.
674 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
675 the linux kernel header. This ensures that dump always has the
676 latest version of this file. Patch submitted by
677 Andreas Dilger <adilger@turbolinux.com>.
679 4. Report any filesystem present in either /etc/fstab with a
680 positive passno or /etc/dumpdates in dump -w output.
681 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
683 5. Fixed the looping problem in dump introduced in the
684 previous version.
686 6. Changed the -B option of dump to limit the size of
687 _compressed_ output per volume if compression is on.
688 Patch contributed by Helmut Jarausch
689 <jarausch@igpm.rwth-aachen.de>. Note however that, since
690 it is impossible to predict the size of the compressed
691 data before doing the compression, the -B limit is a bit
692 conservative.
694 7. Fixed a bug in reading the operator typed file/tape path for
695 the next volume in restore.
697 8. Implemented a "-F script" option to restore which permits the
698 user to specify a script which will be launched at the
699 beginning of each tape, useful for automatic programming of
700 tape changers for example. See the restore man page for the
701 script parameters and return codes.
703 9. Small fix for the QFA routines provided by Uwe Gohlke
704 <uwe@ugsoft.de>, and some recommendations for QFA uses in
705 the man pages.
707 10. Fixed the multivolume restoring where making a mistake
708 to the 'Mount next tape' prompt caused several blocks to
709 be lost.
711 11. Enhanced the -e option of dump to take as a parameter a
712 comma separated list of inode numbers.
714 12. Added the -E option to dump which specify a file containing
715 inode numbers to exclude from the dump.
717 13. Fixed the compressed multi-volume dump + restore.
719 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
720 ==================================================================
722 1. Made dump report the number of blocks dumped per volume.
723 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
725 2. Fix a bug in dump where waiting too much at the 'change volumes'
726 question caused the volume to be rewritten. Thanks to
727 Graham Reed <greed@users.sourceforge.net> for reporting the
728 bug and providing a patch.
730 3. Added a compression option to dump, contributed by Jerry
731 Peters <gapeters@worldnet.att.net>.
733 WARNING: the compression works only on files/pipes and
734 drives supporting variable block size.
736 WARNING2: tapes created using the compression feature are
737 incompatible with the standard dump tape format, and a
738 version of restore >= 0.4b22 will be needed for extraction.
740 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
741 architectures. Thanks to Paul Slootman <paul@debian.org> for
742 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
743 it upstream.
745 5. Many cleanups (CPP defines, const char warnings, check of
746 ext2fs COMPAT flags, time_t cleanups, added libext2 version
747 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
749 6. Made --prefix option work in configure. All the install path
750 are now based on the configure parameters.
752 7. Added the Quick File Access mode in dump/restore, contributed
753 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
754 a file tape position for each inode, and this file is used by
755 restore (if called with parameter Q and the filename)
756 to directly position the tape at the file restore is currently
757 working on. This saves hours when restoring single files from
758 large backups, saves the tapes and the drive's head. Use
759 --enable-qfa option of configure to compile in the QFA support.
761 8. Added the possibility to dump several files and directories
762 in a single invocation of dump. Thanks to Uwe Gohlke
763 <uwe@ugsoft.de> for implementing this option.
765 9. Fixed the dumping and restoring of files with holes
766 (files ending with holes didn't get dumped or restored
767 correctly).
769 10. Fixed a socket descriptor leak in dump, which leaved opened
770 3 file descriptors per dump process (and there is one dump
771 process per tape).
773 11. Fixed dump large file system support, by explicit use of
774 open64/lseek64/etc functions (explicit use needed because
775 e2fsprogs libraries don't behave well when compiled with
778 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
779 ======================================================================
781 1. Fixed some bugs in the dump -w|-W logic introduced by
782 the previous version. Thanks to Andreas Dilger
783 <adilger@turbolinux.com> for his help on this one.
785 2. Fixed again a compilation problem when using old e2fs
786 headers (filesystem label related). Thanks to many users
787 who reported this stupid error.
789 3. Fixed a build problem on old lib5 based systems dealing with
790 _PATH_MNTTAB being undefined. Thanks to John Adams
791 <johna@onevista.com> for reporting the bug.
793 4. Improved the error detection in restore when creating
794 the temporary files in TMPDIR. Now restore will corectly
795 report a 'no space left on device' error instead of
796 strange errors which could imply an archive corruption.
797 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
798 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
800 5. Added the throughput information in dump's progression
801 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
802 for the patch.
804 6. Use libext2fs's inode scanning routines, which are particularly
805 robust when dealing with errors (especially when having some
806 bad blocks on the filesystem). Many thanks to Colin
807 <colin@horizon.com> for reporting the bug, investigating
808 the issues, finding a workaround, writing the patch and
809 fully testing it... (of course, if this patch will break
810 anything, he is to blame for :-)).
812 7. Made dump and restore LFS aware. Dump can dump a filesystem
813 containing large files, generate a large file on output and
814 restore can restore them. This feature is not enabled by
815 default, use --enable-largefile option of configure to enable
816 it (you need to have a LFS aware glibc though). Thanks to
817 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
818 and to Theodore T'so <tytso@valinux.com> for his always
819 useful thoughts.
821 8. Made dump ask upon a tape write error if it should rewrite
822 the current volume (assume this is a bad tape) or if it should
823 assume an end-of-tape condition (useful for tape drives which
824 behaves badly at the end of the tape). Thanks to Andreas
825 Dilger <adilger@turbolinux.com> for the suggestion.
827 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
828 =======================================================================
830 1. Fixed a small compilation problem due to a change
831 in the definintion of the struct sigaction in
832 glibc 2.0 / libc5. Thanks to Gunther Schlegel
833 <schlegel@riege.de> for reporting the bug and to
834 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
835 a fix.
837 2. Modified the label and uuid analysis in order to be
838 self-contained (independant of kernel/libc headers). This
839 should fix the compile with older kernel/libc headers and
840 will preserve the functionality. Thanks to Bernhard Erdmann
841 <bernhard.erdmann@gmx.de> for reporting the bug.
843 3. The 'exclude inode' option, if applied to a directory
844 inode, excludes now correctly all the leaves of this
845 directory. Thanks to John R. Dennison
846 <gerdesas@users.sourceforge.net> for reporting the bug.
848 4. Fixed the '-e' option to disable the possibility
849 to exclude the root inode (which causes the dumps to
850 be unreadable by restore). Prevented array overflow
851 when multiple -e options are used.
853 5. Fixed dump to correctly interpret a filesystem argument
854 which is a mountpoint and it is not an absolute path
855 (as specified in the fstab). Thanks to Bernhard R. Erdmann
856 <be@berdmann.de> for reporting the bug.
858 6. Made dump able to backup files larger than 2 GB. Note that
859 dump still doesn't cope with files larger than 4 GB.
861 7. Restore the real uid and gid before invoking an external
862 RSH program (big hole when dump or restore is suid root!).
864 8. Get the values for the default tape device and the location
865 of fstab file from the system headers. Thanks to
866 Andreas Dilger <adilger@turbolinux.com> for the patch.
868 9. Made dump -w|-W report all recognized filesystems
869 present in either /etc/fstab or /etc/dumpdates, and present
870 the list in the same order as in fstab file. Thanks
871 to Andreas Dilger <adilger@turbolinux.com> for the patch.
873 10. Made dump's -a (automatic end of tape detection) the
874 default. Specifying one of -B, -c, -d or -s options will
875 override the EOM detection. Thanks to Andreas Dilger
876 <adilger@turbolinux.com> for the patch.
878 11. Save the ext2 filesystem label into the dump volume label.
879 Specifying a volume label on the command line (-L option)
880 will override this feature. Thanks to Andreas Dilger
881 <adilger@turbolinux.com> for the patch.
883 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
884 =====================================================================
886 1. Fixed the signal handling in dump (which I broke in 0.4b17)
887 which was causing several strange dump failures (dump
888 hanged or segmentation faults etc).
890 2. Specified the default blocksize in the dump man page.
892 3. Changed two info messages of restore to be written on stdout
893 instead of stderr in order to leave stderr only for errors
894 or warnings. Thanks to Stephen Atwell
895 <satwell@urbana.css.mot.com> for the suggestion.
897 4. Corrected an off by one calculation which prevented
898 dumping a directory having a 1 character name length.
899 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
900 for reporting the bug.
902 5. Reinforce dump in order to allow only level 0 and no
903 -u option when dumping a subdirectory, and document
904 this in the man page. Thanks to Bernhard Erdmann
905 <bernhard.erdmann@gmx.de> for reporting the bug.
907 6. Fixed a small harmless bug in restore which caused
908 in some conditions hard links to be restored several
909 times (therefore generation some warning messages).
910 Thanks to Kenneth Porter <shiva@well.com> for
911 reporting the bug.
913 7. Updated the RPM spec file to the latest RedHat version,
914 providing FHS packaging and other cosmetic changes.
915 You will need RPM version >= 3.0.5 in order to build the RPM.
917 8. Updated the configure script to check for libtermcap
918 before checking for libreadline, because we need this
919 library in order to compile the readline support.
921 9. Made dump understand the LABEL= and UUID= notation
922 both in /etc/fstab and on the command line. Note that
923 you will need the /proc filesystem in order to use
924 these notations. Thanks to Erik Troan <ewt@redhat.com>
925 for providing the patch.
927 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
928 ===================================================================
930 1. Fixed a potential buffer overflow in restore. Thanks
931 to Stan Bubrouski <satan@fastdial.net> for reporting
932 the bug.
934 2. Fixed a readline-related bug which prevented
935 'cat DUMPFILE | restore -i -f -' from working. Thanks
936 to Charles Karney <karney@users.sourceforge.net>
937 for the bug report.
939 3. Changed a few "panic" into "exit", causing restore to
940 be more stable against some attacks (like the last one
941 reported on Bugtraq, although the last version of restore
942 was not vulnerable - just dumped core). Thanks to
943 Andreas Hasenack <andreas@conectiva.com.br> for reporting
944 the bugs to me.
946 4. Removed the suid-root bit on dump and restore in the
947 default build (and generated RPMs). It should be safer
948 now. Users who need the suid-root capabilities in order
949 to make network backups should read first the man page
950 of dump and enable the suid bit manually.
952 5. Added -ltermcap to the compile parameters for restore
953 when using readline support, in order to make the compile
954 process work on some systems (SuSE ?). Thanks to
955 Patrik Schindler <poc@pocnet.net> for reporting the bug.
957 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
958 ====================================================================
960 1. The -F script is called now *only* at the end of a tape,
961 not at the end of the dump. Thanks to Christian Haul
962 <haul@informatik.tu-darmstadt.de> for the bug report.
964 Normally, the device name and the current volume number
965 are passed on the command line of the script. However,
966 if you want the old style script (with no arguments
967 passed on the command line), you can enable it in
968 configure with the --enable-oldstylefscript.
970 2. Use posix signal handling to preserve dump functionality
971 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
972 the patch.
974 3. Made the exit code of restore in case of a 'C'ompare
975 command reflect the result of the compare. An exit status
976 of 0 means the dump archive is correct, 1 means tape errors,
977 2 means that some files were modified. Thanks to Kenneth Porter
978 <shiva@well.com> for the suggestion.
980 4. Made (finally) quotes work in interactive restore.
982 5. Small fixes in order to allow dump to compile with a
983 really old e2fsprogs version. Thanks to Ian Zimmerman
984 <itz@speakeasy.org> for the bug report.
986 6. Add GNU readline capabilities to interactive restore.
987 Use configure's --enable-readline flag to enable this feature.
988 Thanks to Patrik Schindler <poc@pocnet.net> for the
989 suggestion.
991 7. Do the compare on the fly in restore 'C' mode (this will
992 allow not to exhaust the available /tmp space when
993 ccmparing large files). Thanks to Kenneth Porter
994 <shiva@well.com> for the suggestion.
996 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
997 ====================================================================
999 1. Fixed some several duplicate 'DUMP: DUMP:' in the
1000 output of dump.
1002 2. Corrected the estimation of blocks to dump. Note that
1003 this is still wrong for large directory inodes, due
1004 to the size of a BSD directory entry on the tape
1005 (estimation is based on the size of the ext2 directory,
1006 which is smaller because it doesn't contain the null
1007 character at the end).
1009 3. Made dump report the total number of MB written to
1010 tape at the end of run. Thanks to W. Reilly Cooley
1011 <wcooley@nakedape.cc> for the patch.
1013 4. Added the -X option to restore giving the possibility
1014 to read the names of the files to be extracted/listed
1015 from a text file (in addition of the command line).
1016 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
1017 patch.
1019 5. Added the device name and the current volume number
1020 as arguments to the end of tape script (dump -F option).
1022 6. Made the multi-volume dump work again (after having
1023 broken it in 0.4b15).
1025 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
1026 ===================================================================
1028 1. Added a prompt command in interactive restore mode. Thanks
1029 to Andreas Dilger <adilger@home.com> for the patch.
1031 2. Fixed a buffer overflow problem in dump (caused by
1032 not checking the size of the filesystem parameter).
1033 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
1034 reporting this on Bugtraq (and to several dump users
1035 who forwarded me his mail).
1037 3. Added the '-F script' option to dump in order to
1038 launch a script at the end of each tape (to be used
1039 with a tape changer, or to notify the sysadmin by
1040 pager etc.).
1042 4. Fixed a bug in restore compare code caused by the changes
1043 I made in 0.4b14.
1045 5. Fixed the treatment of options using the old BSD syntax
1046 in both dump and restore.
1048 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
1049 =======================================================================
1051 1. Fixed a bug in dump which may have caused invalid deleted
1052 directories to be dumped out if they were deleted after the
1053 mapping pass. This could occure on active filesystem and lead
1054 to heap corruption (causing dump malfunction in many possible ways).
1055 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
1056 bug and submitting the patch.
1058 2. Fixed the handling of the filesystem parameter in dump. A
1059 '/mnt/disk' parameter caused the disk contents to be dumped,
1060 but a '/mnt/disk/' parameter caused the mountpoint directory
1061 to be dumped (generally an empty directory).
1063 3. Improved the output of dump in order to tell which directory
1064 it is currently dumping (when dumping a subtree).
1066 4. Added the '-e' exclude inode option to dump. Thanks to
1067 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
1069 5. Added a REPORTING-BUGS file in order to provide a guide
1070 on how to correctly report a bug in dump/restore.
1072 6. Fixed a restore problem when restoring a hard link to an inode
1073 having the immutable or append-only attribute set. Thanks to
1074 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
1076 7. Fixed a compatibility problem between dumps made on little
1077 endian machines (the format was correct) and big endian
1078 machines (the format was incorrect). This fix break the
1079 compatibility with the older dumps made on big endian
1080 machines (sparc, m86k, ppc etc). For the first time in
1081 linux dump's history (I believe), the dumps made by *BSD,
1082 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
1083 so interchangeable. Thanks to Rob Cermak
1084 <cermak@ahab.rutgers.edu> for submitting the bug and
1085 helping me test the fix.
1087 8. Fixed the way dump reports the remaining percent/time, if
1088 the number of blocks actually dumped exceeds the estimated
1089 number of blocks. Thanks to Jean-Paul van der Jagt
1090 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
1092 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
1093 ======================================================================
1095 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
1096 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
1098 2. Removed the requirement to build the RPM as root from the
1099 spec file. Thanks to Christian Weisgerber
1100 <naddy@mips.rhein-neckar.de> for submitting this.
1102 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
1103 causing dump to block when dumping really huge filesystems.
1104 Many thanks to Patrik Schindler <poc@pocnet.net> for
1105 helping me find this bug.
1107 4. Fixed the treatment for an interrupt signal when dump access
1108 the remote tape through RSH. Thanks to Christian Weisgerber
1109 <naddy@mips.rhein-neckar.de> for providing the patch.
1111 5. Fixed a bug which was causing dump/restore to display
1112 garbage characters instead of the remote host name.
1114 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
1115 =====================================================================
1117 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
1118 <kukuk@suse.de> for submitting the patch.
1120 2. Fix for the exit code when using the size estimate option of
1121 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
1122 submitting the patch.
1124 3. Handle EINTR in atomical reads/writes in dump, which was causing
1125 dump to fail on some systems. Thanks to Eric Jergensen
1126 <eric@dvns.com> for reporting the bug and submitting the patch.
1128 4. Handle more than 16 characters for the device names in dumpdates.
1129 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
1130 tracking down the problem and proposing the solution.
1132 5. Fixed a bug in dump which prevented the creation of the
1133 dumpdates file when doing a 0-level dump without already
1134 having a dumpdates file. Thanks to Patrik Schindler
1135 <poc@pocnet.net> for reporting the bug.
1137 6. Changed the way dump 'S' flag reports the size estimate
1138 from number of blocks into bytes (making it compatible
1139 with the Solaris version, and simplifying things for
1140 amanda users). Thanks to Jason L Tibbitts III
1141 <tibbs@math.uh.edu> for reporting the bug.
1143 7. Fixed a compatibility problem in linux/alpha dump tape format.
1144 Now the linux/alpha dump are (again) compatible with the
1145 other dump formats. But this breaks compatibility with
1146 older dumps made on alpha. Thanks to Mike Tibor
1147 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
1149 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
1150 ======================================================================
1152 1. Added a '--enable-kerberos' to configure.
1154 2. Added a 'S' option to dump which determines the amount of space
1155 that is needed to perform the dump without actually doing it, similar
1156 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
1157 <cermak@ahab.rutgers.edu>.
1159 3. Added a 'M' multi-volume option to dump and restore which enables
1160 dumping to multiple files (useful when dumping to an ext2
1161 partition to make several dump files in order to bypass the 2GB
1162 file size limitation). The 'f' argument is treated as a prefix and
1163 the output files will be named <prefix>001, <prefix>002 etc. With
1164 the 'M' flag, restore automatically selects the right file without
1165 asking to enter a new tape each time.
1167 4. Fixed a memory leak which was causing dump to grow very big
1168 (270MB when dumping a 10GB filesystem...). Thanks to Jason
1169 Fearon <jasonf@netrider.org.au> for reporting the bug.
1171 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
1172 ======================================================================
1174 1. Make configure test if the system glob routines support
1175 extended syntax (ALTDIRFUNC). If not, use the internal glob
1176 routines instead of system ones. Thanks to Bernhard Sadlowski
1177 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
1178 and helping me resolve this and other minor libc5 compiling
1179 glitches.
1181 2. Fix a problem when dumping a ext2fs with the 'filetype'
1182 feature enabled. Thanks to Patrick J. LoPresti
1183 <patl@cag.lcs.mit.edu> for reporting the bug and to
1184 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
1186 3. Made the nodump flag work on directories. A directory which
1187 has the nodump flag gets never dumped, regardless of its
1188 contents.
1190 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
1191 which allows dump on an active ext3 filesystem. However, this
1192 is a "quick and dirty" patch which enables backup of an ext3
1193 filesystem through the ext2 compatibility (by ignoring the
1194 NEEDS_RECOVERY bit). The journal file is not recognized and
1195 it is dumped (it should not).
1197 5. Test the superblock compatibility flags when dumping, in order
1198 to be sure that we know how to deal with specific features.
1200 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
1201 ====================================================================
1203 1. Use lchown instead of chown, fixing a possible security problem
1204 when restoring symlinks (a malicious user could use this
1205 to deliberately corrupt the ownership of important system files).
1206 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
1207 this and providing the patch.
1209 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
1210 ====================================================================
1212 1. Put dump sources under CVS, added Id tags in all files so
1213 one can use 'ident' on binary files.
1215 2. Added the dump/restore version in the usage text so one can
1216 easily verify the version he is using.
1218 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
1219 a va_start/va_end problem on linux-ppc (always call va_start
1220 va_end in pairs each time we use a vararg function).
1222 4. Added again the DT_* constants because old libc does not
1223 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
1224 for submitting the bug report.
1226 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
1227 this should enable dumping big (huge) filesystems.
1229 6. Added the RSH environment variable in order to be able to
1230 use a rsh replacement like ssh when doing remote backups (and
1231 bypass the security limitations of rcmd). Now you can do remote
1232 backups without being root (or making dump setuid root).
1234 7. Modified again the way dumpdates works. For incremental dumps,
1235 we need to read dumpdates even if we are not using 'u' option.
1236 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
1237 this should work.
1239 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
1240 ===================================================================
1242 1. Removed the 'k' flag from the restore 'about' text if kerberos
1243 was not compiled in.
1245 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
1246 to them (fsetflags takes a char*, setflags an open fd!).
1248 3. (f)setflags is called only if the flags aren't empty. If the
1249 file is a special file, a warning is printed, because changing
1250 flags implies opening the device. Normally, a special file
1251 should not have any flag... (Debian bug #29775, patch provided
1252 by Abhijit Dasgupta <abhijit@ans.net>).
1254 4. Made possible to dump a file system not mentioned in /etc/fstab.
1255 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
1257 5. Changed the default behaviour to not create dumpdates
1258 unless 'u' option is specified. Removed the old "debian-patch"
1259 which provided the same thing. (Debian bug #38136, #33818).
1261 6. Removed all those dump*announce, since they were getting old...
1263 7. Added warning messages if dumpdates does not exist and
1264 when an inferior level dump does not exist (except for a level 0
1265 dump).
1267 8. Debugged the glob calls in interactive mode: restore used a
1268 dirent struct which was different from the /usr/include/dirent.h
1269 one (this used to work, is it a glibc2 change?), so none of the
1270 compat glob (which used /usr/include/dirent.h) or the system glob
1271 worked. Restore use now the system dirent (and the system
1272 DT_* constants), which are compatible with BSD ones.
1274 9. Added a configure flag (--with-dumpdatespath) to specify
1275 the location of dumpdates file. By default, it is
1276 /etc/dumpdates.
1278 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1279 included the current date/version in man pages.
1281 11. Corrected the estimation of remaining time when
1282 the operator doesn't change the tapes quickly enough. This
1283 was an old bug, I thought I corrected it, and discovered
1284 that in fact it was corrected in two different places, so
1285 the results canceled each other...
1287 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1288 ===================================================================
1290 1. Integrated multiple patches from RedHat, Debian and SuSE:
1292 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1293 isn't specified.
1294 - several fixes in the man pages.
1295 - update the default tape device to /dev/st0.
1296 - many updates for Linux Alpha (byte ordering, size_t etc).
1297 - buffer overruns.
1298 - use environment variable for TMPDIR (instead of /tmp).
1299 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1300 - workaround egcs bug (RedHat bugs #4281 and #2989).
1301 - wire $(OPT) throughout Makefile's.
1303 2. Upgrade the dump revision to 1, making possible to dump filesystems
1304 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1306 3. Fix some compile warnings, prototype all functions.
1308 4. Use glibc err/glob instead of internal compatibility
1309 routines (only if available).
1311 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1312 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1314 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1315 ======================================================================
1317 1. Integrated the changes from FreeBSD-3.1-RELEASE
1318 (mostly bug fixes, buffer overruns, dump has now an "automatic
1319 tape length calculation" flag, dump/restore can use kerberos now
1320 (this is NOT tested), use environment variables for TAPE and
1321 RMT etc.).
1323 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1325 3. Corrected a bug in restore when using 'C' option with multi-volumes
1326 tapes (files splited accros two tapes give "size changed" errors
1327 when comparing).
1329 4. Corrected the long standing bug when dumping multiple tapes.
1330 This works for me, needs further testing.
1332 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1333 ====================================================================
1335 1. Dump now runs correctly on kernels 2.1.x
1336 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1338 Changes between versions 0.4b2 and 0.4b3
1339 ========================================
1341 1. Use realpath() if available
1343 2. Report statistics
1345 Changes between versions 0.4b1 and 0.4b2
1346 ========================================
1348 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1349 the patch)
1351 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1353 3. Fixed configure.in to correctly handle --enable-rmt
1355 Changes between versions 0.3 and 0.4b1
1356 ======================================
1358 1. Integrated the changes from 4.4BSD-Lite2
1360 2. Integrated the patches from Debian and Red Hat
1362 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1364 4. Changed dump to use the Ext2fs library to get block addresses. This
1365 should solve the endianness problem on SparcLinux.
1367 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1368 distribution's one) to use autoconf
1370 6. Fixed a few minor bugs
1372 Changes between versions 0.2e and 0.2f
1373 ======================================
1375 1. Added the creation of named pipes (fifos) in restore.
1377 2. Added the -N flag in the restore manual page.
1379 3. Added the file kernel.patch which contains the llseek() optimization
1380 patch for 1.2.x kernels.
1382 4. Fixed a bug in the restoration of symbolic links: owner and group were
1383 not restored.
1385 5. Integrated some changes from FreeBSD 2.2.
1387 6. Added a call to ftruncate() after restoring each file to restore
1388 correctly files ending by a hole.
1390 Changes between versions 0.2d and 0.2e
1391 ======================================
1393 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1394 with owner = root (0) and group = root/wheel/whatever (0).
1396 Changes between versions 0.2c and 0.2d
1397 ======================================
1399 1. Dump is now able to backup 2GB+ filesystems.
1401 2. Dump and restore can now be linked as static binaries.
1403 Changes between versions 0.2b and 0.2c
1404 ======================================
1406 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1408 Changes between versions 0.2a and 0.2b
1409 ======================================
1411 1. Really fixed the bug that I should have corrected in 0.2a.
1413 2. Enabled optimization again.
1415 Changes between versions 0.2 and 0.2a
1416 =====================================
1418 1. Disabled the optimization during compilation.
1420 Changes between versions 0.1 and 0.2
1421 ====================================
1423 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1424 the fs_type field (actually, I modified the file fstab.c to make it
1425 use the mntent functions).
1427 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1428 made by dump should be readable by the BSD restore and backups made
1429 by the BSD dump should be readable by restore. Unfortunately, this
1430 means that the dump format has changed between version 0.1 and version
1431 0.2 :-(
1433 3. Dump is now able to backup a subtree, it is no longer limited to whole
1434 filesystems like the BSD version.
1436 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1437 than 2 GB.
1439 Changes between versions 0.0 and 0.1
1440 ====================================
1442 1. Now create links rdump and rrestore during the `make install' step.
1444 2. Linux port specific bugs added to the manual pages
1446 3. Incorrect estimation of the number of tapes blocks fixed when doing
1447 incremental backups.
1449 4. Better ls-like format in restore in interactive mode.