1c3488f7a2d8bb474778da168fa7f9164c80d383
[dump.git] / CHANGES
1 $Id: CHANGES,v 1.277 2005/06/08 13:24:08 stelian Exp $
2
3 Changes between versions 0.4b40 and 0.4b41 (released ???????????)
4 =================================================================
5
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.
9
10 2. Fix some issues with restoration of EA on big endian
11 platforms.
12
13 Changes between versions 0.4b39 and 0.4b40 (released May 2, 2005)
14 =================================================================
15
16 1. Changed restore to emit warnings (instead of emitting a fatal
17 error) if a file (or a directory) is unavailable for a
18 comparision (if the user doesn't have the necessary permissions
19 to access it for example). Thanks to Kenneth Porter
20 <shiva@sewingwitch.com> for the bug report.
21
22 2. Re-done the 'do not save directory entries to non-dumped inodes
23 (excluded from dump)' feature. The previous implementation
24 worked well for excluded directories but not for regular files.
25 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
26 report.
27
28 3. Fixed a bug in dump where the tape size was miscalculated when
29 the user used -d/-s to specify the tape characteristics. Thanks
30 to Philip Goisman <goisman@physics.arizona.edu> for reporting
31 the bug.
32
33 4. Fixed another bug introduced in restore with the hashtree
34 implementation. This one caused restore to stop saying
35 "removenode: non-empty directory" in some cases.
36
37 5. Added support for dumping and restoring ext2/3 extended
38 attributes (EA), like the access control lists (ACL) or
39 the security labels used by SELinux.
40
41 Changes between versions 0.4b38 and 0.4b39 (released January 21, 2005)
42 ======================================================================
43
44 1. The newly added dump_on_cd_3 example was buggy, replace it
45 with an updated version from Andrew Basterfield
46 <bob@cemetery.homeunix.org>.
47
48 2. Made restore to chdir() back into the initial directory when
49 dumping core while aborting a comparision operation. The
50 previous behaviour was to write the corefile at the root of
51 the directory being compared, which could very well be
52 read only and preventing the corefile generation. Thanks
53 to Kenneth Porter <shiva@sewingwitch.com> for the bug report.
54
55 3. Silenced the failure to call fgetflags() when comparing an
56 entry which has no ext2 attributes (as in lsattr()).
57
58 4. Fix a brown paper bug in restore -C which broke restore and
59 caused modifications on the filesystem being compared
60 (directories containing a file with the same name as the
61 directory get renamed to RSTTMP...). Thanks to Kenneth Porter
62 <shiva@sewingwitch.com> for finding the bug and helping me
63 reproduce it.
64
65 5. Made restore -C force the -N flag (no writing allowed on
66 the disk) in order to prevent more bugs like the above one.
67
68 Changes between versions 0.4b37 and 0.4b38 (released January 7, 2005)
69 =====================================================================
70
71 1. Fix a couple of troff syntax bugs in the man pages.
72 Thanks to Eric Raymond <esr@thyrsus.com> for the patch.
73
74 2. Made restore use either libncurses or libtermcap, depending
75 on which one is available at configure time.
76
77 3. Fixed restore negative size display bug when comparing a
78 dump containing files over 2GB. Thanks to Steve Bonds
79 <sbonds@users.sourceforge.net> for the bug report.
80
81 4. Do not save directory entries to non-dumped inodes
82 (excluded from dump). This will eliminate the 'missing
83 file' warnings when doing 'restore -C'.
84
85 5. Fix dump crash when backuping a huge (2TB) filesystem,
86 due to a bogus calculation on the inode map size.
87 Thanks to Kevin B. Haines <K.B.Haines@rl.ac.uk> for
88 submitting the bug and testing the fix.
89
90 6. Fix a problem in restore where the final \0 in the symbolic
91 link names could have been lost, generating corrupt filenames.
92 Thanks to Kyle Wilson <kyle.wilson@amd.com> for reporting the
93 bug.
94
95 7. Implemented a hash list for the directory names in restore.
96 The linear list used before caused problems in interactive
97 restores when dealing with directories having thousands of
98 entries. Thanks to Brian Ristuccia <bristuccia@starentnetworks.com>
99 for reporting the bug.
100
101 8. Improved restore -C, this time including the directory
102 attributes into the comparision.
103
104 9. Made restore understand tapes containing EA/ACLs (which will
105 be dumped by the next version of dump). In this version
106 extended attributes on the tape are ignored, for full EA/ACL
107 support wait for the next version or try the experimental EA
108 patch.
109
110 Changes between versions 0.4b36 and 0.4b37 (released July 7, 2004)
111 ==================================================================
112
113 1. Added the --enable-staticz configure option which enables
114 dump and restore to be linked with static versions of
115 libz and libbz2 (and dynamic versions of all other libraries).
116 This will make Debian users happy, because libz and libbz2
117 were the only needed libraries living in /usr, all the
118 others live in /lib. In case of system emergency, it is
119 better not to have to rely on an extra filesystem.
120 Thanks to Bdale Garbee <bdale@gag.com> for the suggestion.
121
122 2. Fix compilation on (at least the Linux Debian port to) AMD64.
123 (<ext2fs/ext2_types.h> defines some types (__s64 and __u64)
124 that are also defined by <linux/types.h> (<asm/types.h>) and
125 they conflict).
126
127 3. Make dump's reading of the dumpdates file a bit more robust,
128 preventing dump from crashing when the dumpdates file has
129 been modified by hand.
130
131 4. Fixed some offset calculations in dump code which could
132 lead to "bread lseek errors" on large filesystems. Thanks
133 to Bruce Lowekamp <lowekamp@users.sourceforge.net> for
134 reporting this bug and debugging the issue.
135
136 5. Made dump use the blkid library when searching for devices
137 by label or uuid instead of dump's own routines.
138
139 6. Corrected a bug in dump where a wrong LABEL=... line in
140 /etc/fstab could prevent dump from dumping unrelated
141 filesystems. Thanks to Bruce Lowekamp
142 <lowekamp@users.sourceforge.net> for reporting the bug.
143
144 Changes between versions 0.4b35 and 0.4b36 (released April 21, 2004)
145 ====================================================================
146
147 1. Fixed dump compilation with old gcc versions. Thanks to
148 Mike Castle <dalgoda@users.sourceforge.net> for the patch.
149
150 2. Fixed some warnings (howmany, roundup, powerof2 redefined)
151 when compiling against a recent glibc version.
152
153 3. Fixed a bug in restore preventing the read of a dump tape
154 written with Solaris 7 ufsdump. Thanks to Patrick Higgins
155 <phiggins@transzap.com> for reporting the bug and providing
156 the test case.
157
158 4. Changed dump to enable the creation of volumes of different
159 sizes in a single run (make -B accept a list of values).
160 Patch contributed by Florian Zumbiehl <florz@gmx.de>.
161
162 5. Use the glibc provided minor() and major() macros instead
163 of our own bitmask implementation. This should be safe for
164 when the major/minor namespace will migrate to 32 bits.
165 Thanks to Zhang Jun <zhangjun@nanjing-fnst.com> for reporting
166 the bug.
167
168 6. Made explicit in the dump man page that dump will not create
169 a remote file, it will only write to an already existing one.
170
171 7. Another try at making size estimates better again.
172
173 8. Put back the inconditional running of the end-of-tape
174 script which was changed in 0.434 to be run only when -M
175 or multiple -f were NOT specified. Some users rely on this
176 feature even when it is combined with -M/-f.
177
178 9. Fixed restore when restoring huge backups (where rstdir...
179 temporary files are over 2GB). Thanks to
180 Raphael Jaffey <rjaffey@artic.edu> for reporting this,
181 debugging the issue and making the patch.
182
183 10. Made restore understand FreeBSD UFS2 tapes. Thanks to
184 David <vrtin@users.sourceforge.net> for submitting the bug
185 and providing a test case.
186
187 11. Made dump work with an arbitrary integer level (no more 10
188 levels only limitation). Thanks to Matthew
189 <msvincen@midway.uchicago.edu> for the patch.
190
191 Changes between versions 0.4b34 and 0.4b35 (released December 21, 2003)
192 =======================================================================
193
194 1. Added a note in the dump man page saying that the default
195 blocksize can be 32 if -d is used with a high density argument.
196 Thanks to Antonios Christofides <A.Christofides@itia.ntua.gr>
197 for the patch.
198
199 2. Fixed configure to correctly understand CPPFLAGS, CFLAGS,
200 LDFLAGS environment variables. Thanks to Arcady Genkin
201 <antipode@users.sourceforge.net> for reporting the bug.
202
203 3. Made -e/-E options of dump accept an unlimited number of inodes
204 to be excluded, rather than a hardcoded maximum. Thanks to
205 Dietrich Rothe <d-rothe@users.sourceforge.net> for the patch.
206
207 4. Updated the autoconf system to 2.50. Forced the -D_BSD_SOURCE
208 and -D_USE_BSD_SIGNAL defines in configure in order to solve
209 64bit build problems because quad_t is redifined with a
210 different signature. Thanks to Mike Harris <mharris@redhat.com>
211 for reporting this bug.
212
213 5. Made restore build on Solaris, making possible to
214 restore Linux's "enhanced" tapes. Thanks to Uwe Gohlke
215 <uwe@ugsoft.de> for the patch.
216
217 6. Made an extension in the dump tape format capable of saving
218 MacOSX specific inode extensions. Uwe Gohlke <uwe@ugsoft.de>
219 wrote the extension and contributed the restore code back
220 into this codebase. The same extension mechanism will be
221 used in the future to save ACLs...
222
223 7. Made rmt work correctly with regard to QFA and local/remote
224 files and tapes. The remote access will however work only
225 when the dump provided rmt version is used. If you want to
226 use another rmt server, please do not use the QFA feature.
227 Thanks to Uwe Gohlke <uwe@ugsoft.de> for the patch.
228
229 Changes between versions 0.4b33 and 0.4b34 (released April 18, 2003)
230 ====================================================================
231
232 1. Fixed the INSTALL file to reflect the actual install paths.
233 Thanks to David Ronis <ronis@ronispc.chem.mcgill.ca> for
234 reporting the bug.
235
236 2. Fixed the configure script to only check for headers presence
237 instead of trying to compile them. This should fix issues
238 with old build environments. Thanks to Kari Lempiainen
239 <kari@funky.pp.fi> for reporting the bug.
240
241 3. Fixed restore to correctly ignore sockets when comparing
242 a dump (as socket cannot be properly restored anyway). Thanks
243 to Gunther Reiszig <gunther@mit.edu> for reporting the bug.
244
245 4. Fixed restore to correctly access the archive file (-A argument)
246 even when using a remote tape. Thanks to Fabrice Bellet
247 <fabrice@bellet.info> for reporting the bug.
248
249 5. Fixed (again) handling of long (largefile) seeks in rmt.
250 Thanks to Fabrice Bellet <fabrice@bellet.info> for reporting
251 the bug.
252
253 6. Fixed restore corner case when dealing with large block sizes
254 dump is able to write now (-b 1024). Thanks to Fabrice Bellet
255 <fabrice@bellet.info> for reporting the bug.
256
257 7. Fixed a bug preventing dump to access a filesystem having
258 a label exactly 16 bytes in length. Thanks to <minduni@ti-edu.ch>
259 for reporting the bug.
260
261 8. Made dump store dump dates using explicit timezones, fixing a
262 problem with incremental dumps if the timezone is modified
263 between the dumps. Thanks to Philippe Troin <phil@fifi.org> for
264 the bug report and the patch.
265
266 9. Fixed a bug encountered when dumping individual files (not full
267 filesystems or directories) and dangling symbolic links happen
268 to be in the list of files. For as far as dump is concerned,
269 dangling symbolic links are allowed, and are dumped as is.
270 Thanks to Jin-su Ahn <jsahn@ee.snu.ac.kr> for reporting the
271 bug and providing the fix.
272
273 10. Fixed open and creation modes and permissions for QFA and
274 table-of-contents files in dump and restore. Thanks to
275 Philippe Troin <phil@fifi.org> for the patch.
276
277 11. Fixed the archive file descriptor handling enabling it to be 0.
278 This can happen in some cases when shell redirections are used.
279 Thanks to Philippe Troin <phil@fifi.org> for the patch.
280
281 12. Delayed the opening of archive file until after suid had been
282 dropped (fixing a possible security issue if dump is suid).
283 Thanks to Philippe Troin <phil@fifi.org> for the patch.
284
285 13. Fixed the 'S' command handling in the rmt client part.
286 Thanks to Philippe Troin <phil@fifi.org> for the patch.
287
288 14. Modified the end-of-tape script handling to print out statistics
289 (and stop the timer) before launching the eot script. Also, the eot
290 script does not get run anymore when using -M (which makes sense) or
291 when multiple tapes are listed on the command line
292 (-f tape0,tape1,tapen) (which also makes sense).
293 Thanks to Philippe Troin <phil@fifi.org> for the patch.
294
295 15. Relicensed dump/restore under the 'revised' BSD license, as per
296 ftp://ftp.cs.berkeley.edu/ucb/4bsd/README.Impt.License.Change.
297
298 16. Added LZO compression to dump. This new compression method has
299 the advantage of being super fast, thus not killing tape streaming
300 on slow machines. Thanks to Helmut Jarausch
301 <jarausch@igpm.rwth-aachen.de> for the patch and to
302 Markus Oberhumer <markus@oberhumer.com> for giving special permission
303 to include his miniLZO project (GPL licensed) in dump/restore.
304
305 17. Some small buffer overruns fixes in rmt. Thanks to Antonomasia
306 <ant@notatla.demon.co.uk> for reporting the bugs.
307
308 18. Added a special rmt version which can do encryption when writing
309 to tape. Read examples/encrypted_rmt/README for details on
310 how to enable and configure it. Thanks to Ken Lalonde
311 <ken@globalremit.com> for the patch.
312
313 19. Made dump work with 2.5 kernel end of tape early warning semantics.
314 Thanks to Kai Makisara <Kai.Makisara@kolumbus.fi> for the patch.
315
316 20. Fixed a bug which caused dump -w|-W not to work anymore, because
317 the fs_freq and fs_passno fields in /etc/mtab are always set
318 to 0 0. Thanks to Trent Piepho <xyzzy@speakeasy.org> for
319 reporting the bug.
320
321 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
322 =======================================================================
323
324 1. Added a note in the restore man page clarifying the question
325 'set the permissions on the current directory ?' asked by
326 restore at the end of treatment in -i and -x modes.
327
328 2. Fixed the endianess issues when reading compressed tapes.
329 Thanks to Dark Force <daq4th@users.sourceforge.net> for
330 reporting this bug and providing test cases.
331
332 3. Fixed the "ACL won't be dumped" warning message (which showed
333 an extra, unrelated error message). Thanks to Dragan Krnic
334 <dkrnic@lycos.com> for reporting this bug.
335
336 4. Made dump look first into /etc/mtab, then into /etc/fstab
337 when searching for filesystem paths. Also fixed some problems
338 caused by binding mounts or multiple block device mounts.
339 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
340 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
341 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
342 and #160305, Sourceforge bugs #618699 and #679832).
343
344 5. Made dump's -I option accept the value '0' meaning all the
345 read errors will be ignored. This can be useful when running
346 dump from unattended sessions (like cron jobs). Thanks to
347 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
348
349 6. Fixed the output of dump to indicate 'blocks' instead of
350 'tape blocks' in the various messages (blocks are always
351 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
352 made it clearly print the current blocksize at the start of
353 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
354 the suggestions.
355
356 7. Made rmt understand long (largefiles) seeks.
357
358 8. Fixed build with very old versions of libext2fs, where
359 EXT2_FT_* constants were undefined.
360
361 9. Made dump accept the dumpdates path on the command line
362 (-D file option) instead of using only the hardcoded one.
363 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
364 suggestion.
365
366 10. Enabled rmt, LFS, readline, QFA options by default in
367 ./configure. Updated the configure process (new versions
368 of config.guess, config.sub etc).
369
370 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
371 =======================================================================
372
373 1. Changed dump to use fcntl(F_SETLK) style locking instead
374 of flock() when locking the dumpdates file. With the old
375 locking scheme, a local user having read rights on the
376 dumpdates file could be able to do a Denial of Service attack
377 on dump. In order to lock the dumpdates file with the new
378 scheme, the user would need to have write access on the file.
379 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
380 reporting the bug (originally a bugtraq post).
381
382 2. Fixed interactive 'ls' which caused spurious errors warnings
383 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
384 <jorgen_ostling@users.sourceforge.net> for reporting this
385 bug.
386
387 3. Fixed dump's estimate when dealing with sparse inodes.
388
389 4. Modified dump to allow setting a blocksize bigger than 32kB
390 (raised the limit to 1024kB), because newer hardware needs
391 this for proper operation. Thanks to Dirk Traenapp
392 <dtraenapp@users.sourceforge.net> for reporting this.
393
394 5. Fixed a bug causing Dump to stop and report an error if an
395 inode number in the exclude file was followed by some amount
396 of whitespace. Thanks to Jeffrey Sofferin
397 <sofferin@users.sourceforge.net> for reporting this bug.
398
399 6. Fixed a bug which caused restore, in some particular cases,
400 to ask some 'scary' questions and leave a bunch of RSTTMP
401 directories behind when restoring incremental tapes. Thanks
402 to Philippe Troin <phil@fifi.org> for reporting this bug and
403 providing the test cases.
404
405 7. Changed the wording when inodes are excluded from dump:
406 replaced 'Added inode 7 to exclude list' with
407 'Excluding inode 7 (resize inode) from dump', as suggested
408 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
409
410 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
411 ===================================================================
412
413 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
414 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
415 <eros@lamel.bo.cnr.it> for reporting the bug.
416
417 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
418 ===================================================================
419
420 1. Made dump print out the ext2 logical block number in case of a read
421 error, which can be given as an argument to debugfs' ncheck command,
422 in order to find out the affected inode. Added note about this
423 usage in dump's man page.
424
425 2. Fixed a problem in restore when reading tapes written on big
426 endian machines with very old versions of dump. The patch was
427 contributed by George Helffrich <george@geology.bristol.ac.uk>.
428
429 3. Fixed the tape length calculation when using large tapes
430 and compression. Thanks to Georg Lippold
431 <g_lippold@sourceforge.net> for reporting the bug.
432
433 4. Added a new set of examples in dump_on_cd_2 directory, based
434 on dump_on_cd examples but somewhat enhanced, supporting DVD
435 media, and localized in english and german. Thanks to
436 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
437
438 5. Made dump save 32 bit UID/GID. Internally, this was achieved
439 by switching from the old BSD inode format to the new BSD
440 inode format, which means that the tape format was changed.
441 However, since all restore versions out there should
442 transparently support both inode formats, the change should
443 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
444 reporting the bug.
445
446 6. Fixed a lot of warnings in the code shown when compiling
447 with 'gcc -W'. Thanks to Matthias Andree
448 <matthias.andree@stud.uni-dortmund.de> for reporting this.
449
450 7. Fixed a small markup bug in the dump man page. Thanks to
451 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
452 patch.
453
454 8. Rewrote entirely the man pages using the tmac.an macro
455 package (Linux man page format) instead of the original BSD
456 format. They should be now cleaner and easier to modify.
457
458 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
459 ==================================================================
460
461 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
462 non Linux clients were misinterpreted. The description of the problem
463 (incompatible numbering in Linux mtio opcodes) is documented at
464 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
465 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
466 providing an excellent, cross-platform replacement for rmt in his
467 star package.
468
469 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
470 which caused restore to fail when the inode maps were not entirely
471 contained on the first volume (when using really small volumes or
472 when dumping a huge number of inodes).
473
474 3. Fixed a problem in dump, where files in subdirectories of directories
475 that have the nodump flag set (or being excluded with -e/-E) were
476 sometimes incorrectly dumped (instead of being excluded). The original
477 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
478 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
479 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
480 help testing the patch.
481
482 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
483 which makes one able to pipe the output of dump, by the net, to
484 a remote CD-burner server.
485
486 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
487 and use GNU's symbolic syntax over rmt instead of numerical values
488 to assure multiple platform compatibility.
489
490 6. Documented the -d option in restore.
491
492 7. Added a -v (verbose) mode to dump. For now it just prints the number
493 of the inode being dumped, but this could evolve in future versions
494 to include interesting debugging output.
495
496 8. Added a -o flag to restore, which automatically restores the current
497 directory permissions in -i or -x mode, without asking the operator
498 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
499 and Peter Samuel <peters@e-smith.com>.
500
501 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
502 ====================================================================
503
504 1. Fixed a bug in the QFA feature which could have caused
505 a bad tape positionning on restore (causing extra delays in
506 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
507
508 2. Added a small note in the dump man page specifying that
509 there should be no white space between the option letter and
510 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
511
512 3. Made restore work with older versions of the readline library,
513 by checking in configure for several extended symbols. Restore
514 can now be compiled with a readline as old as the 2.0 release
515 (though it may be a good idea to upgrade it to a more recent
516 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
517 reporting the build failures.
518
519 4. Fixed a performance problem with the QFA file creation in
520 dump, which made unnecessary seeks on the tape slowing down
521 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
522 reporting this issue.
523
524 5. Removed the inclusion of some kernel headers in the dump
525 source, which prevented the compile in some kernel/glibc
526 headers/architecture combination. Thanks to Bdale Garbee
527 <bdale@gag.com> for reporting the bug.
528
529 6. Added the appropriate error message when dump fails to
530 open the output file for writing. Thanks to Amith Varghese
531 <amithv@yahoo.com> for reporting this bug.
532
533 7. Made restore able to understand large Solaris ufsdump tapes
534 (containing inodes bigger than 4194304). Sun have introduced
535 an "extension" to the dump tape format when dealing with
536 those inodes, which was uncorrectly handled by Linux restore.
537 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
538 providing a test case.
539
540 8. Added the -m parameter to dump which optimises the output for
541 inodes having been changed but not modified since the last dump
542 ('changed' and 'modified' have the meaning defined in stat(2)).
543 For those inodes, dump will save only the metadata, instead of
544 saving the entire inode contents. Inodes which are either
545 directories or have been modified since the last dump are saved
546 in a regular way. Uses of this flag must be consistent, meaning
547 that either every dump in an incremental dump set have the flag,
548 or no one has it.
549
550 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
551 =======================================================================
552
553 1. Fixed behaviour of dump when exceeding resource limits
554 (SIGXFSZ treatment).
555
556 2. Added the -L flag to restore to allow the user to specify a
557 maximal allowed number of miscompares when using restore
558 with the -C option to check the backup.
559
560 3. Detailed the manual entry for the -N option of restore.
561
562 4. Added the -a flag to restore to make able doing unattended
563 restores in -i or -x mode (automatically walks through the
564 multiple dump volumes).
565
566 5. Extended the QFA mode to work with local files and/or
567 remote tapes and files. This way, restore can know in advance
568 the tape number and the offset for the inodes to extract and
569 can minimize the extraction time by seeking directly to the
570 good tape/offset.
571
572 6. Added the -A <archive> option to both dump and restore,
573 which makes dump to archive a dump table-of-contents in
574 the specified file to be used by restore to determine
575 whether a file is in the dump file that is being restored.
576 (the archive file syntax is also compatible with the
577 Solaris ufsdump generated one).
578
579 7. Small fix in restore making it able to read some (broken ?)
580 Solaris ufsdump tapes.
581
582 8. Fixed dump to correctly recognise the root filesystem when using
583 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
584 for reporting this bug.
585
586 9. Added the -P <file> option to restore to create a
587 Quick File Access file from an already made dump. Patch
588 contributed by Uwe Gohlke <uwe@ugsoft.de>.
589
590 10. Made restore compile and run on Solaris, making it a
591 possible replacement for the standard ufsrestore. Port was
592 contributed by Uwe Gohlke <uwe@ugsoft.de>.
593
594 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
595 =====================================================================
596
597 1. Added a set of backup scripts from Eugenio Diaz
598 <getnito@yahoo.com> in the examples section. It features
599 automatic (cron based) full and incremental dumping of
600 several filesystems on a separate filesystem.
601
602 2. Fixed a off-by-one miscalculation which disabled dumping
603 a one letter subdirectory. Thanks to Chris Siebenmann
604 <cks@utcc.utoronto.ca> for reporting the bug.
605
606 3. Fixed several restore bugs occuring when trying to
607 restore some missing files on the tape. Thanks to Chris
608 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
609
610 4. Fixed --with-ldopts configure argument passing, installing from
611 a separate object directory, makefile cleanups contributed
612 by <splite@purdue.edu>.
613
614 5. Fix a bug which could caused, in some conditions, the highest
615 number inode of a filesystem, to not be dumped. Many thanks
616 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
617 chase this bug.
618
619 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
620 =======================================================================
621
622 1. Added a mini howto from Patrick Walsh in the examples
623 directory.
624
625 2. Minor man pages syntax corrections. Thanks to
626 Chris Danis <screechco@home.com> for reporting the bugs.
627
628 3. Added a script from David B. Peterson <dave@toppledwagon.com>
629 to the examples section. It features dumping several
630 filesystems to a remote tape drive upon ssh.
631
632 4. Added a patch provided by Richard Jones <rich@annexia.org>
633 which allows BRADEMAX (number of read errors tolerated by
634 dump) to be adjusted using the -I option.
635
636 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
637 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
638 patch.
639
640 6. Add the -l option to restore to specify if, when doing a
641 remote restore, the file used to access the backup is a
642 regular file (the defaults being a tape drive). Restore needs
643 to know this information when reading a remote compressed
644 dump. Previously, this information was autodetected, but
645 the autodetection code fails (with ioctl: Inappropriate ioctl
646 for device) when using a non Linux remote box. Thanks to
647 many users and especially to Eros Albertazzi
648 <eros@lamel.bo.cnr.it> for reporting this.
649
650 7. Found a workaroung for the dump deadlock problem (3 childs
651 stuck in pause(), father in read()). The workaround seems
652 to work for me and several beta-testers. If it doesn't work
653 for you, please report back.
654
655 8. Updated the RPM spec file (BuildPrereq, URL etc).
656
657 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
658 ========================================================================
659
660 1. Fixed the permissions of a newly created QFA file by dump.
661
662 2. Cleaned up the source of dump (the external variables
663 definition was a complete mess, making possible to have
664 objects overlap).
665
666 3. Fixed restore to use the full tape volume path when doing
667 a compare (since it changes the working directory to the
668 filesystem being compared in the process).
669
670 4. Added the -q option to dump which makes dump abort
671 whenever it needs operator attention. It should be
672 easier to use dump in scripts with this option.
673
674 5. Detect the use of incompatible options to dump and
675 refuse them (like -a and -B options together).
676
677 6. Added bzip2 compression to dump/restore (use option -j level
678 to select it). Note that you will need a restore version
679 >= 0.4b24 in order to restore a bzip2 compressed dump.
680 The same warning as for the zlib compression applies:
681 the tape format of a bzip2 dump is not compatible with the
682 original BSD tape format.
683
684 7. Fixed a overflow problem in dump corrupting the dump when
685 very large files were encountered. Thanks to Vladimir Nadvornik
686 for the bug report.
687
688 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
689 kernel buffer/page cache before starting the dump, helping
690 a bit those who use dump on mounted filesystems. Thanks to
691 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
692 for this suggestion.
693
694 9. Updated the RPM spec file following the RedHat changes
695 (dynamically linked binaries now in /usr/sbin etc).
696
697 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
698 which enables restore to recognise multi volume compressed dumps
699 done on CD. Included his scripts for dump (which pipe the dump
700 output directly into cdrecord) and restore. There is now
701 possible to to backups to CD on the fly!
702
703 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
704 ===================================================================
705
706 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
707 Marcus Meissner (Caldera International Security Dept.).
708
709 2. Implement the Sun rmt extended protocol. Patch submitted
710 by Ian Gordon <iangordon@users.sourceforge.net>.
711
712 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
713 the linux kernel header. This ensures that dump always has the
714 latest version of this file. Patch submitted by
715 Andreas Dilger <adilger@turbolinux.com>.
716
717 4. Report any filesystem present in either /etc/fstab with a
718 positive passno or /etc/dumpdates in dump -w output.
719 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
720
721 5. Fixed the looping problem in dump introduced in the
722 previous version.
723
724 6. Changed the -B option of dump to limit the size of
725 _compressed_ output per volume if compression is on.
726 Patch contributed by Helmut Jarausch
727 <jarausch@igpm.rwth-aachen.de>. Note however that, since
728 it is impossible to predict the size of the compressed
729 data before doing the compression, the -B limit is a bit
730 conservative.
731
732 7. Fixed a bug in reading the operator typed file/tape path for
733 the next volume in restore.
734
735 8. Implemented a "-F script" option to restore which permits the
736 user to specify a script which will be launched at the
737 beginning of each tape, useful for automatic programming of
738 tape changers for example. See the restore man page for the
739 script parameters and return codes.
740
741 9. Small fix for the QFA routines provided by Uwe Gohlke
742 <uwe@ugsoft.de>, and some recommendations for QFA uses in
743 the man pages.
744
745 10. Fixed the multivolume restoring where making a mistake
746 to the 'Mount next tape' prompt caused several blocks to
747 be lost.
748
749 11. Enhanced the -e option of dump to take as a parameter a
750 comma separated list of inode numbers.
751
752 12. Added the -E option to dump which specify a file containing
753 inode numbers to exclude from the dump.
754
755 13. Fixed the compressed multi-volume dump + restore.
756
757 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
758 ==================================================================
759
760 1. Made dump report the number of blocks dumped per volume.
761 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
762
763 2. Fix a bug in dump where waiting too much at the 'change volumes'
764 question caused the volume to be rewritten. Thanks to
765 Graham Reed <greed@users.sourceforge.net> for reporting the
766 bug and providing a patch.
767
768 3. Added a compression option to dump, contributed by Jerry
769 Peters <gapeters@worldnet.att.net>.
770
771 WARNING: the compression works only on files/pipes and
772 drives supporting variable block size.
773
774 WARNING2: tapes created using the compression feature are
775 incompatible with the standard dump tape format, and a
776 version of restore >= 0.4b22 will be needed for extraction.
777
778 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
779 architectures. Thanks to Paul Slootman <paul@debian.org> for
780 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
781 it upstream.
782
783 5. Many cleanups (CPP defines, const char warnings, check of
784 ext2fs COMPAT flags, time_t cleanups, added libext2 version
785 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
786
787 6. Made --prefix option work in configure. All the install path
788 are now based on the configure parameters.
789
790 7. Added the Quick File Access mode in dump/restore, contributed
791 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
792 a file tape position for each inode, and this file is used by
793 restore (if called with parameter Q and the filename)
794 to directly position the tape at the file restore is currently
795 working on. This saves hours when restoring single files from
796 large backups, saves the tapes and the drive's head. Use
797 --enable-qfa option of configure to compile in the QFA support.
798
799 8. Added the possibility to dump several files and directories
800 in a single invocation of dump. Thanks to Uwe Gohlke
801 <uwe@ugsoft.de> for implementing this option.
802
803 9. Fixed the dumping and restoring of files with holes
804 (files ending with holes didn't get dumped or restored
805 correctly).
806
807 10. Fixed a socket descriptor leak in dump, which leaved opened
808 3 file descriptors per dump process (and there is one dump
809 process per tape).
810
811 11. Fixed dump large file system support, by explicit use of
812 open64/lseek64/etc functions (explicit use needed because
813 e2fsprogs libraries don't behave well when compiled with
814 FILE_OFFSET_BITS=64).
815
816 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
817 ======================================================================
818
819 1. Fixed some bugs in the dump -w|-W logic introduced by
820 the previous version. Thanks to Andreas Dilger
821 <adilger@turbolinux.com> for his help on this one.
822
823 2. Fixed again a compilation problem when using old e2fs
824 headers (filesystem label related). Thanks to many users
825 who reported this stupid error.
826
827 3. Fixed a build problem on old lib5 based systems dealing with
828 _PATH_MNTTAB being undefined. Thanks to John Adams
829 <johna@onevista.com> for reporting the bug.
830
831 4. Improved the error detection in restore when creating
832 the temporary files in TMPDIR. Now restore will corectly
833 report a 'no space left on device' error instead of
834 strange errors which could imply an archive corruption.
835 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
836 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
837
838 5. Added the throughput information in dump's progression
839 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
840 for the patch.
841
842 6. Use libext2fs's inode scanning routines, which are particularly
843 robust when dealing with errors (especially when having some
844 bad blocks on the filesystem). Many thanks to Colin
845 <colin@horizon.com> for reporting the bug, investigating
846 the issues, finding a workaround, writing the patch and
847 fully testing it... (of course, if this patch will break
848 anything, he is to blame for :-)).
849
850 7. Made dump and restore LFS aware. Dump can dump a filesystem
851 containing large files, generate a large file on output and
852 restore can restore them. This feature is not enabled by
853 default, use --enable-largefile option of configure to enable
854 it (you need to have a LFS aware glibc though). Thanks to
855 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
856 and to Theodore T'so <tytso@valinux.com> for his always
857 useful thoughts.
858
859 8. Made dump ask upon a tape write error if it should rewrite
860 the current volume (assume this is a bad tape) or if it should
861 assume an end-of-tape condition (useful for tape drives which
862 behaves badly at the end of the tape). Thanks to Andreas
863 Dilger <adilger@turbolinux.com> for the suggestion.
864
865 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
866 =======================================================================
867
868 1. Fixed a small compilation problem due to a change
869 in the definintion of the struct sigaction in
870 glibc 2.0 / libc5. Thanks to Gunther Schlegel
871 <schlegel@riege.de> for reporting the bug and to
872 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
873 a fix.
874
875 2. Modified the label and uuid analysis in order to be
876 self-contained (independant of kernel/libc headers). This
877 should fix the compile with older kernel/libc headers and
878 will preserve the functionality. Thanks to Bernhard Erdmann
879 <bernhard.erdmann@gmx.de> for reporting the bug.
880
881 3. The 'exclude inode' option, if applied to a directory
882 inode, excludes now correctly all the leaves of this
883 directory. Thanks to John R. Dennison
884 <gerdesas@users.sourceforge.net> for reporting the bug.
885
886 4. Fixed the '-e' option to disable the possibility
887 to exclude the root inode (which causes the dumps to
888 be unreadable by restore). Prevented array overflow
889 when multiple -e options are used.
890
891 5. Fixed dump to correctly interpret a filesystem argument
892 which is a mountpoint and it is not an absolute path
893 (as specified in the fstab). Thanks to Bernhard R. Erdmann
894 <be@berdmann.de> for reporting the bug.
895
896 6. Made dump able to backup files larger than 2 GB. Note that
897 dump still doesn't cope with files larger than 4 GB.
898
899 7. Restore the real uid and gid before invoking an external
900 RSH program (big hole when dump or restore is suid root!).
901
902 8. Get the values for the default tape device and the location
903 of fstab file from the system headers. Thanks to
904 Andreas Dilger <adilger@turbolinux.com> for the patch.
905
906 9. Made dump -w|-W report all recognized filesystems
907 present in either /etc/fstab or /etc/dumpdates, and present
908 the list in the same order as in fstab file. Thanks
909 to Andreas Dilger <adilger@turbolinux.com> for the patch.
910
911 10. Made dump's -a (automatic end of tape detection) the
912 default. Specifying one of -B, -c, -d or -s options will
913 override the EOM detection. Thanks to Andreas Dilger
914 <adilger@turbolinux.com> for the patch.
915
916 11. Save the ext2 filesystem label into the dump volume label.
917 Specifying a volume label on the command line (-L option)
918 will override this feature. Thanks to Andreas Dilger
919 <adilger@turbolinux.com> for the patch.
920
921 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
922 =====================================================================
923
924 1. Fixed the signal handling in dump (which I broke in 0.4b17)
925 which was causing several strange dump failures (dump
926 hanged or segmentation faults etc).
927
928 2. Specified the default blocksize in the dump man page.
929
930 3. Changed two info messages of restore to be written on stdout
931 instead of stderr in order to leave stderr only for errors
932 or warnings. Thanks to Stephen Atwell
933 <satwell@urbana.css.mot.com> for the suggestion.
934
935 4. Corrected an off by one calculation which prevented
936 dumping a directory having a 1 character name length.
937 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
938 for reporting the bug.
939
940 5. Reinforce dump in order to allow only level 0 and no
941 -u option when dumping a subdirectory, and document
942 this in the man page. Thanks to Bernhard Erdmann
943 <bernhard.erdmann@gmx.de> for reporting the bug.
944
945 6. Fixed a small harmless bug in restore which caused
946 in some conditions hard links to be restored several
947 times (therefore generation some warning messages).
948 Thanks to Kenneth Porter <shiva@well.com> for
949 reporting the bug.
950
951 7. Updated the RPM spec file to the latest RedHat version,
952 providing FHS packaging and other cosmetic changes.
953 You will need RPM version >= 3.0.5 in order to build the RPM.
954
955 8. Updated the configure script to check for libtermcap
956 before checking for libreadline, because we need this
957 library in order to compile the readline support.
958
959 9. Made dump understand the LABEL= and UUID= notation
960 both in /etc/fstab and on the command line. Note that
961 you will need the /proc filesystem in order to use
962 these notations. Thanks to Erik Troan <ewt@redhat.com>
963 for providing the patch.
964
965 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
966 ===================================================================
967
968 1. Fixed a potential buffer overflow in restore. Thanks
969 to Stan Bubrouski <satan@fastdial.net> for reporting
970 the bug.
971
972 2. Fixed a readline-related bug which prevented
973 'cat DUMPFILE | restore -i -f -' from working. Thanks
974 to Charles Karney <karney@users.sourceforge.net>
975 for the bug report.
976
977 3. Changed a few "panic" into "exit", causing restore to
978 be more stable against some attacks (like the last one
979 reported on Bugtraq, although the last version of restore
980 was not vulnerable - just dumped core). Thanks to
981 Andreas Hasenack <andreas@conectiva.com.br> for reporting
982 the bugs to me.
983
984 4. Removed the suid-root bit on dump and restore in the
985 default build (and generated RPMs). It should be safer
986 now. Users who need the suid-root capabilities in order
987 to make network backups should read first the man page
988 of dump and enable the suid bit manually.
989
990 5. Added -ltermcap to the compile parameters for restore
991 when using readline support, in order to make the compile
992 process work on some systems (SuSE ?). Thanks to
993 Patrik Schindler <poc@pocnet.net> for reporting the bug.
994
995 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
996 ====================================================================
997
998 1. The -F script is called now *only* at the end of a tape,
999 not at the end of the dump. Thanks to Christian Haul
1000 <haul@informatik.tu-darmstadt.de> for the bug report.
1001
1002 Normally, the device name and the current volume number
1003 are passed on the command line of the script. However,
1004 if you want the old style script (with no arguments
1005 passed on the command line), you can enable it in
1006 configure with the --enable-oldstylefscript.
1007
1008 2. Use posix signal handling to preserve dump functionality
1009 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
1010 the patch.
1011
1012 3. Made the exit code of restore in case of a 'C'ompare
1013 command reflect the result of the compare. An exit status
1014 of 0 means the dump archive is correct, 1 means tape errors,
1015 2 means that some files were modified. Thanks to Kenneth Porter
1016 <shiva@well.com> for the suggestion.
1017
1018 4. Made (finally) quotes work in interactive restore.
1019
1020 5. Small fixes in order to allow dump to compile with a
1021 really old e2fsprogs version. Thanks to Ian Zimmerman
1022 <itz@speakeasy.org> for the bug report.
1023
1024 6. Add GNU readline capabilities to interactive restore.
1025 Use configure's --enable-readline flag to enable this feature.
1026 Thanks to Patrik Schindler <poc@pocnet.net> for the
1027 suggestion.
1028
1029 7. Do the compare on the fly in restore 'C' mode (this will
1030 allow not to exhaust the available /tmp space when
1031 ccmparing large files). Thanks to Kenneth Porter
1032 <shiva@well.com> for the suggestion.
1033
1034 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
1035 ====================================================================
1036
1037 1. Fixed some several duplicate 'DUMP: DUMP:' in the
1038 output of dump.
1039
1040 2. Corrected the estimation of blocks to dump. Note that
1041 this is still wrong for large directory inodes, due
1042 to the size of a BSD directory entry on the tape
1043 (estimation is based on the size of the ext2 directory,
1044 which is smaller because it doesn't contain the null
1045 character at the end).
1046
1047 3. Made dump report the total number of MB written to
1048 tape at the end of run. Thanks to W. Reilly Cooley
1049 <wcooley@nakedape.cc> for the patch.
1050
1051 4. Added the -X option to restore giving the possibility
1052 to read the names of the files to be extracted/listed
1053 from a text file (in addition of the command line).
1054 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
1055 patch.
1056
1057 5. Added the device name and the current volume number
1058 as arguments to the end of tape script (dump -F option).
1059
1060 6. Made the multi-volume dump work again (after having
1061 broken it in 0.4b15).
1062
1063 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
1064 ===================================================================
1065
1066 1. Added a prompt command in interactive restore mode. Thanks
1067 to Andreas Dilger <adilger@home.com> for the patch.
1068
1069 2. Fixed a buffer overflow problem in dump (caused by
1070 not checking the size of the filesystem parameter).
1071 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
1072 reporting this on Bugtraq (and to several dump users
1073 who forwarded me his mail).
1074
1075 3. Added the '-F script' option to dump in order to
1076 launch a script at the end of each tape (to be used
1077 with a tape changer, or to notify the sysadmin by
1078 pager etc.).
1079
1080 4. Fixed a bug in restore compare code caused by the changes
1081 I made in 0.4b14.
1082
1083 5. Fixed the treatment of options using the old BSD syntax
1084 in both dump and restore.
1085
1086 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
1087 =======================================================================
1088
1089 1. Fixed a bug in dump which may have caused invalid deleted
1090 directories to be dumped out if they were deleted after the
1091 mapping pass. This could occure on active filesystem and lead
1092 to heap corruption (causing dump malfunction in many possible ways).
1093 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
1094 bug and submitting the patch.
1095
1096 2. Fixed the handling of the filesystem parameter in dump. A
1097 '/mnt/disk' parameter caused the disk contents to be dumped,
1098 but a '/mnt/disk/' parameter caused the mountpoint directory
1099 to be dumped (generally an empty directory).
1100
1101 3. Improved the output of dump in order to tell which directory
1102 it is currently dumping (when dumping a subtree).
1103
1104 4. Added the '-e' exclude inode option to dump. Thanks to
1105 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
1106
1107 5. Added a REPORTING-BUGS file in order to provide a guide
1108 on how to correctly report a bug in dump/restore.
1109
1110 6. Fixed a restore problem when restoring a hard link to an inode
1111 having the immutable or append-only attribute set. Thanks to
1112 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
1113
1114 7. Fixed a compatibility problem between dumps made on little
1115 endian machines (the format was correct) and big endian
1116 machines (the format was incorrect). This fix break the
1117 compatibility with the older dumps made on big endian
1118 machines (sparc, m86k, ppc etc). For the first time in
1119 linux dump's history (I believe), the dumps made by *BSD,
1120 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
1121 so interchangeable. Thanks to Rob Cermak
1122 <cermak@ahab.rutgers.edu> for submitting the bug and
1123 helping me test the fix.
1124
1125 8. Fixed the way dump reports the remaining percent/time, if
1126 the number of blocks actually dumped exceeds the estimated
1127 number of blocks. Thanks to Jean-Paul van der Jagt
1128 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
1129
1130 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
1131 ======================================================================
1132
1133 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
1134 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
1135
1136 2. Removed the requirement to build the RPM as root from the
1137 spec file. Thanks to Christian Weisgerber
1138 <naddy@mips.rhein-neckar.de> for submitting this.
1139
1140 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
1141 causing dump to block when dumping really huge filesystems.
1142 Many thanks to Patrik Schindler <poc@pocnet.net> for
1143 helping me find this bug.
1144
1145 4. Fixed the treatment for an interrupt signal when dump access
1146 the remote tape through RSH. Thanks to Christian Weisgerber
1147 <naddy@mips.rhein-neckar.de> for providing the patch.
1148
1149 5. Fixed a bug which was causing dump/restore to display
1150 garbage characters instead of the remote host name.
1151
1152 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
1153 =====================================================================
1154
1155 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
1156 <kukuk@suse.de> for submitting the patch.
1157
1158 2. Fix for the exit code when using the size estimate option of
1159 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
1160 submitting the patch.
1161
1162 3. Handle EINTR in atomical reads/writes in dump, which was causing
1163 dump to fail on some systems. Thanks to Eric Jergensen
1164 <eric@dvns.com> for reporting the bug and submitting the patch.
1165
1166 4. Handle more than 16 characters for the device names in dumpdates.
1167 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
1168 tracking down the problem and proposing the solution.
1169
1170 5. Fixed a bug in dump which prevented the creation of the
1171 dumpdates file when doing a 0-level dump without already
1172 having a dumpdates file. Thanks to Patrik Schindler
1173 <poc@pocnet.net> for reporting the bug.
1174
1175 6. Changed the way dump 'S' flag reports the size estimate
1176 from number of blocks into bytes (making it compatible
1177 with the Solaris version, and simplifying things for
1178 amanda users). Thanks to Jason L Tibbitts III
1179 <tibbs@math.uh.edu> for reporting the bug.
1180
1181 7. Fixed a compatibility problem in linux/alpha dump tape format.
1182 Now the linux/alpha dump are (again) compatible with the
1183 other dump formats. But this breaks compatibility with
1184 older dumps made on alpha. Thanks to Mike Tibor
1185 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
1186
1187 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
1188 ======================================================================
1189
1190 1. Added a '--enable-kerberos' to configure.
1191
1192 2. Added a 'S' option to dump which determines the amount of space
1193 that is needed to perform the dump without actually doing it, similar
1194 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
1195 <cermak@ahab.rutgers.edu>.
1196
1197 3. Added a 'M' multi-volume option to dump and restore which enables
1198 dumping to multiple files (useful when dumping to an ext2
1199 partition to make several dump files in order to bypass the 2GB
1200 file size limitation). The 'f' argument is treated as a prefix and
1201 the output files will be named <prefix>001, <prefix>002 etc. With
1202 the 'M' flag, restore automatically selects the right file without
1203 asking to enter a new tape each time.
1204
1205 4. Fixed a memory leak which was causing dump to grow very big
1206 (270MB when dumping a 10GB filesystem...). Thanks to Jason
1207 Fearon <jasonf@netrider.org.au> for reporting the bug.
1208
1209 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
1210 ======================================================================
1211
1212 1. Make configure test if the system glob routines support
1213 extended syntax (ALTDIRFUNC). If not, use the internal glob
1214 routines instead of system ones. Thanks to Bernhard Sadlowski
1215 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
1216 and helping me resolve this and other minor libc5 compiling
1217 glitches.
1218
1219 2. Fix a problem when dumping a ext2fs with the 'filetype'
1220 feature enabled. Thanks to Patrick J. LoPresti
1221 <patl@cag.lcs.mit.edu> for reporting the bug and to
1222 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
1223
1224 3. Made the nodump flag work on directories. A directory which
1225 has the nodump flag gets never dumped, regardless of its
1226 contents.
1227
1228 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
1229 which allows dump on an active ext3 filesystem. However, this
1230 is a "quick and dirty" patch which enables backup of an ext3
1231 filesystem through the ext2 compatibility (by ignoring the
1232 NEEDS_RECOVERY bit). The journal file is not recognized and
1233 it is dumped (it should not).
1234
1235 5. Test the superblock compatibility flags when dumping, in order
1236 to be sure that we know how to deal with specific features.
1237
1238 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
1239 ====================================================================
1240
1241 1. Use lchown instead of chown, fixing a possible security problem
1242 when restoring symlinks (a malicious user could use this
1243 to deliberately corrupt the ownership of important system files).
1244 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
1245 this and providing the patch.
1246
1247 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
1248 ====================================================================
1249
1250 1. Put dump sources under CVS, added Id tags in all files so
1251 one can use 'ident' on binary files.
1252
1253 2. Added the dump/restore version in the usage text so one can
1254 easily verify the version he is using.
1255
1256 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
1257 a va_start/va_end problem on linux-ppc (always call va_start
1258 va_end in pairs each time we use a vararg function).
1259
1260 4. Added again the DT_* constants because old libc does not
1261 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
1262 for submitting the bug report.
1263
1264 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
1265 this should enable dumping big (huge) filesystems.
1266
1267 6. Added the RSH environment variable in order to be able to
1268 use a rsh replacement like ssh when doing remote backups (and
1269 bypass the security limitations of rcmd). Now you can do remote
1270 backups without being root (or making dump setuid root).
1271
1272 7. Modified again the way dumpdates works. For incremental dumps,
1273 we need to read dumpdates even if we are not using 'u' option.
1274 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
1275 this should work.
1276
1277 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
1278 ===================================================================
1279
1280 1. Removed the 'k' flag from the restore 'about' text if kerberos
1281 was not compiled in.
1282
1283 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
1284 to them (fsetflags takes a char*, setflags an open fd!).
1285
1286 3. (f)setflags is called only if the flags aren't empty. If the
1287 file is a special file, a warning is printed, because changing
1288 flags implies opening the device. Normally, a special file
1289 should not have any flag... (Debian bug #29775, patch provided
1290 by Abhijit Dasgupta <abhijit@ans.net>).
1291
1292 4. Made possible to dump a file system not mentioned in /etc/fstab.
1293 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
1294
1295 5. Changed the default behaviour to not create dumpdates
1296 unless 'u' option is specified. Removed the old "debian-patch"
1297 which provided the same thing. (Debian bug #38136, #33818).
1298
1299 6. Removed all those dump*announce, since they were getting old...
1300
1301 7. Added warning messages if dumpdates does not exist and
1302 when an inferior level dump does not exist (except for a level 0
1303 dump).
1304
1305 8. Debugged the glob calls in interactive mode: restore used a
1306 dirent struct which was different from the /usr/include/dirent.h
1307 one (this used to work, is it a glibc2 change?), so none of the
1308 compat glob (which used /usr/include/dirent.h) or the system glob
1309 worked. Restore use now the system dirent (and the system
1310 DT_* constants), which are compatible with BSD ones.
1311
1312 9. Added a configure flag (--with-dumpdatespath) to specify
1313 the location of dumpdates file. By default, it is
1314 /etc/dumpdates.
1315
1316 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1317 included the current date/version in man pages.
1318
1319 11. Corrected the estimation of remaining time when
1320 the operator doesn't change the tapes quickly enough. This
1321 was an old bug, I thought I corrected it, and discovered
1322 that in fact it was corrected in two different places, so
1323 the results canceled each other...
1324
1325 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1326 ===================================================================
1327
1328 1. Integrated multiple patches from RedHat, Debian and SuSE:
1329
1330 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1331 isn't specified.
1332 - several fixes in the man pages.
1333 - update the default tape device to /dev/st0.
1334 - many updates for Linux Alpha (byte ordering, size_t etc).
1335 - buffer overruns.
1336 - use environment variable for TMPDIR (instead of /tmp).
1337 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1338 - workaround egcs bug (RedHat bugs #4281 and #2989).
1339 - wire $(OPT) throughout Makefile's.
1340
1341 2. Upgrade the dump revision to 1, making possible to dump filesystems
1342 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1343
1344 3. Fix some compile warnings, prototype all functions.
1345
1346 4. Use glibc err/glob instead of internal compatibility
1347 routines (only if available).
1348
1349 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1350 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1351
1352 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1353 ======================================================================
1354
1355 1. Integrated the changes from FreeBSD-3.1-RELEASE
1356 (mostly bug fixes, buffer overruns, dump has now an "automatic
1357 tape length calculation" flag, dump/restore can use kerberos now
1358 (this is NOT tested), use environment variables for TAPE and
1359 RMT etc.).
1360
1361 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1362
1363 3. Corrected a bug in restore when using 'C' option with multi-volumes
1364 tapes (files splited accros two tapes give "size changed" errors
1365 when comparing).
1366
1367 4. Corrected the long standing bug when dumping multiple tapes.
1368 This works for me, needs further testing.
1369
1370 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1371 ====================================================================
1372
1373 1. Dump now runs correctly on kernels 2.1.x
1374 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1375
1376 Changes between versions 0.4b2 and 0.4b3
1377 ========================================
1378
1379 1. Use realpath() if available
1380
1381 2. Report statistics
1382
1383 Changes between versions 0.4b1 and 0.4b2
1384 ========================================
1385
1386 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1387 the patch)
1388
1389 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1390
1391 3. Fixed configure.in to correctly handle --enable-rmt
1392
1393 Changes between versions 0.3 and 0.4b1
1394 ======================================
1395
1396 1. Integrated the changes from 4.4BSD-Lite2
1397
1398 2. Integrated the patches from Debian and Red Hat
1399
1400 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1401
1402 4. Changed dump to use the Ext2fs library to get block addresses. This
1403 should solve the endianness problem on SparcLinux.
1404
1405 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1406 distribution's one) to use autoconf
1407
1408 6. Fixed a few minor bugs
1409
1410 Changes between versions 0.2e and 0.2f
1411 ======================================
1412
1413 1. Added the creation of named pipes (fifos) in restore.
1414
1415 2. Added the -N flag in the restore manual page.
1416
1417 3. Added the file kernel.patch which contains the llseek() optimization
1418 patch for 1.2.x kernels.
1419
1420 4. Fixed a bug in the restoration of symbolic links: owner and group were
1421 not restored.
1422
1423 5. Integrated some changes from FreeBSD 2.2.
1424
1425 6. Added a call to ftruncate() after restoring each file to restore
1426 correctly files ending by a hole.
1427
1428 Changes between versions 0.2d and 0.2e
1429 ======================================
1430
1431 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1432 with owner = root (0) and group = root/wheel/whatever (0).
1433
1434 Changes between versions 0.2c and 0.2d
1435 ======================================
1436
1437 1. Dump is now able to backup 2GB+ filesystems.
1438
1439 2. Dump and restore can now be linked as static binaries.
1440
1441 Changes between versions 0.2b and 0.2c
1442 ======================================
1443
1444 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1445
1446 Changes between versions 0.2a and 0.2b
1447 ======================================
1448
1449 1. Really fixed the bug that I should have corrected in 0.2a.
1450
1451 2. Enabled optimization again.
1452
1453 Changes between versions 0.2 and 0.2a
1454 =====================================
1455
1456 1. Disabled the optimization during compilation.
1457
1458 Changes between versions 0.1 and 0.2
1459 ====================================
1460
1461 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1462 the fs_type field (actually, I modified the file fstab.c to make it
1463 use the mntent functions).
1464
1465 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1466 made by dump should be readable by the BSD restore and backups made
1467 by the BSD dump should be readable by restore. Unfortunately, this
1468 means that the dump format has changed between version 0.1 and version
1469 0.2 :-(
1470
1471 3. Dump is now able to backup a subtree, it is no longer limited to whole
1472 filesystems like the BSD version.
1473
1474 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1475 than 2 GB.
1476
1477 Changes between versions 0.0 and 0.1
1478 ====================================
1479
1480 1. Now create links rdump and rrestore during the `make install' step.
1481
1482 2. Linux port specific bugs added to the manual pages
1483
1484 3. Incorrect estimation of the number of tapes blocks fixed when doing
1485 incremental backups.
1486
1487 4. Better ls-like format in restore in interactive mode.