BuildRequires: ncurses-devel
[dump.git] / CHANGES
1 $Id: CHANGES,v 1.280 2005/08/20 21:00:47 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 3. Fix restore when the symtab is over 2GB in size.
14
15 4. Made the directory hash indexing an optional feature,
16 accessible by the '-H' option of restore, and disable it
17 by default.
18
19 5. Fixed dump to not include extended attributes information
20 in the toc (archive) file which confused restore -t.
21
22 Changes between versions 0.4b39 and 0.4b40 (released May 2, 2005)
23 =================================================================
24
25 1. Changed restore to emit warnings (instead of emitting a fatal
26 error) if a file (or a directory) is unavailable for a
27 comparision (if the user doesn't have the necessary permissions
28 to access it for example). Thanks to Kenneth Porter
29 <shiva@sewingwitch.com> for the bug report.
30
31 2. Re-done the 'do not save directory entries to non-dumped inodes
32 (excluded from dump)' feature. The previous implementation
33 worked well for excluded directories but not for regular files.
34 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
35 report.
36
37 3. Fixed a bug in dump where the tape size was miscalculated when
38 the user used -d/-s to specify the tape characteristics. Thanks
39 to Philip Goisman <goisman@physics.arizona.edu> for reporting
40 the bug.
41
42 4. Fixed another bug introduced in restore with the hashtree
43 implementation. This one caused restore to stop saying
44 "removenode: non-empty directory" in some cases.
45
46 5. Added support for dumping and restoring ext2/3 extended
47 attributes (EA), like the access control lists (ACL) or
48 the security labels used by SELinux.
49
50 Changes between versions 0.4b38 and 0.4b39 (released January 21, 2005)
51 ======================================================================
52
53 1. The newly added dump_on_cd_3 example was buggy, replace it
54 with an updated version from Andrew Basterfield
55 <bob@cemetery.homeunix.org>.
56
57 2. Made restore to chdir() back into the initial directory when
58 dumping core while aborting a comparision operation. The
59 previous behaviour was to write the corefile at the root of
60 the directory being compared, which could very well be
61 read only and preventing the corefile generation. Thanks
62 to Kenneth Porter <shiva@sewingwitch.com> for the bug report.
63
64 3. Silenced the failure to call fgetflags() when comparing an
65 entry which has no ext2 attributes (as in lsattr()).
66
67 4. Fix a brown paper bug in restore -C which broke restore and
68 caused modifications on the filesystem being compared
69 (directories containing a file with the same name as the
70 directory get renamed to RSTTMP...). Thanks to Kenneth Porter
71 <shiva@sewingwitch.com> for finding the bug and helping me
72 reproduce it.
73
74 5. Made restore -C force the -N flag (no writing allowed on
75 the disk) in order to prevent more bugs like the above one.
76
77 Changes between versions 0.4b37 and 0.4b38 (released January 7, 2005)
78 =====================================================================
79
80 1. Fix a couple of troff syntax bugs in the man pages.
81 Thanks to Eric Raymond <esr@thyrsus.com> for the patch.
82
83 2. Made restore use either libncurses or libtermcap, depending
84 on which one is available at configure time.
85
86 3. Fixed restore negative size display bug when comparing a
87 dump containing files over 2GB. Thanks to Steve Bonds
88 <sbonds@users.sourceforge.net> for the bug report.
89
90 4. Do not save directory entries to non-dumped inodes
91 (excluded from dump). This will eliminate the 'missing
92 file' warnings when doing 'restore -C'.
93
94 5. Fix dump crash when backuping a huge (2TB) filesystem,
95 due to a bogus calculation on the inode map size.
96 Thanks to Kevin B. Haines <K.B.Haines@rl.ac.uk> for
97 submitting the bug and testing the fix.
98
99 6. Fix a problem in restore where the final \0 in the symbolic
100 link names could have been lost, generating corrupt filenames.
101 Thanks to Kyle Wilson <kyle.wilson@amd.com> for reporting the
102 bug.
103
104 7. Implemented a hash list for the directory names in restore.
105 The linear list used before caused problems in interactive
106 restores when dealing with directories having thousands of
107 entries. Thanks to Brian Ristuccia <bristuccia@starentnetworks.com>
108 for reporting the bug.
109
110 8. Improved restore -C, this time including the directory
111 attributes into the comparision.
112
113 9. Made restore understand tapes containing EA/ACLs (which will
114 be dumped by the next version of dump). In this version
115 extended attributes on the tape are ignored, for full EA/ACL
116 support wait for the next version or try the experimental EA
117 patch.
118
119 Changes between versions 0.4b36 and 0.4b37 (released July 7, 2004)
120 ==================================================================
121
122 1. Added the --enable-staticz configure option which enables
123 dump and restore to be linked with static versions of
124 libz and libbz2 (and dynamic versions of all other libraries).
125 This will make Debian users happy, because libz and libbz2
126 were the only needed libraries living in /usr, all the
127 others live in /lib. In case of system emergency, it is
128 better not to have to rely on an extra filesystem.
129 Thanks to Bdale Garbee <bdale@gag.com> for the suggestion.
130
131 2. Fix compilation on (at least the Linux Debian port to) AMD64.
132 (<ext2fs/ext2_types.h> defines some types (__s64 and __u64)
133 that are also defined by <linux/types.h> (<asm/types.h>) and
134 they conflict).
135
136 3. Make dump's reading of the dumpdates file a bit more robust,
137 preventing dump from crashing when the dumpdates file has
138 been modified by hand.
139
140 4. Fixed some offset calculations in dump code which could
141 lead to "bread lseek errors" on large filesystems. Thanks
142 to Bruce Lowekamp <lowekamp@users.sourceforge.net> for
143 reporting this bug and debugging the issue.
144
145 5. Made dump use the blkid library when searching for devices
146 by label or uuid instead of dump's own routines.
147
148 6. Corrected a bug in dump where a wrong LABEL=... line in
149 /etc/fstab could prevent dump from dumping unrelated
150 filesystems. Thanks to Bruce Lowekamp
151 <lowekamp@users.sourceforge.net> for reporting the bug.
152
153 Changes between versions 0.4b35 and 0.4b36 (released April 21, 2004)
154 ====================================================================
155
156 1. Fixed dump compilation with old gcc versions. Thanks to
157 Mike Castle <dalgoda@users.sourceforge.net> for the patch.
158
159 2. Fixed some warnings (howmany, roundup, powerof2 redefined)
160 when compiling against a recent glibc version.
161
162 3. Fixed a bug in restore preventing the read of a dump tape
163 written with Solaris 7 ufsdump. Thanks to Patrick Higgins
164 <phiggins@transzap.com> for reporting the bug and providing
165 the test case.
166
167 4. Changed dump to enable the creation of volumes of different
168 sizes in a single run (make -B accept a list of values).
169 Patch contributed by Florian Zumbiehl <florz@gmx.de>.
170
171 5. Use the glibc provided minor() and major() macros instead
172 of our own bitmask implementation. This should be safe for
173 when the major/minor namespace will migrate to 32 bits.
174 Thanks to Zhang Jun <zhangjun@nanjing-fnst.com> for reporting
175 the bug.
176
177 6. Made explicit in the dump man page that dump will not create
178 a remote file, it will only write to an already existing one.
179
180 7. Another try at making size estimates better again.
181
182 8. Put back the inconditional running of the end-of-tape
183 script which was changed in 0.434 to be run only when -M
184 or multiple -f were NOT specified. Some users rely on this
185 feature even when it is combined with -M/-f.
186
187 9. Fixed restore when restoring huge backups (where rstdir...
188 temporary files are over 2GB). Thanks to
189 Raphael Jaffey <rjaffey@artic.edu> for reporting this,
190 debugging the issue and making the patch.
191
192 10. Made restore understand FreeBSD UFS2 tapes. Thanks to
193 David <vrtin@users.sourceforge.net> for submitting the bug
194 and providing a test case.
195
196 11. Made dump work with an arbitrary integer level (no more 10
197 levels only limitation). Thanks to Matthew
198 <msvincen@midway.uchicago.edu> for the patch.
199
200 Changes between versions 0.4b34 and 0.4b35 (released December 21, 2003)
201 =======================================================================
202
203 1. Added a note in the dump man page saying that the default
204 blocksize can be 32 if -d is used with a high density argument.
205 Thanks to Antonios Christofides <A.Christofides@itia.ntua.gr>
206 for the patch.
207
208 2. Fixed configure to correctly understand CPPFLAGS, CFLAGS,
209 LDFLAGS environment variables. Thanks to Arcady Genkin
210 <antipode@users.sourceforge.net> for reporting the bug.
211
212 3. Made -e/-E options of dump accept an unlimited number of inodes
213 to be excluded, rather than a hardcoded maximum. Thanks to
214 Dietrich Rothe <d-rothe@users.sourceforge.net> for the patch.
215
216 4. Updated the autoconf system to 2.50. Forced the -D_BSD_SOURCE
217 and -D_USE_BSD_SIGNAL defines in configure in order to solve
218 64bit build problems because quad_t is redifined with a
219 different signature. Thanks to Mike Harris <mharris@redhat.com>
220 for reporting this bug.
221
222 5. Made restore build on Solaris, making possible to
223 restore Linux's "enhanced" tapes. Thanks to Uwe Gohlke
224 <uwe@ugsoft.de> for the patch.
225
226 6. Made an extension in the dump tape format capable of saving
227 MacOSX specific inode extensions. Uwe Gohlke <uwe@ugsoft.de>
228 wrote the extension and contributed the restore code back
229 into this codebase. The same extension mechanism will be
230 used in the future to save ACLs...
231
232 7. Made rmt work correctly with regard to QFA and local/remote
233 files and tapes. The remote access will however work only
234 when the dump provided rmt version is used. If you want to
235 use another rmt server, please do not use the QFA feature.
236 Thanks to Uwe Gohlke <uwe@ugsoft.de> for the patch.
237
238 Changes between versions 0.4b33 and 0.4b34 (released April 18, 2003)
239 ====================================================================
240
241 1. Fixed the INSTALL file to reflect the actual install paths.
242 Thanks to David Ronis <ronis@ronispc.chem.mcgill.ca> for
243 reporting the bug.
244
245 2. Fixed the configure script to only check for headers presence
246 instead of trying to compile them. This should fix issues
247 with old build environments. Thanks to Kari Lempiainen
248 <kari@funky.pp.fi> for reporting the bug.
249
250 3. Fixed restore to correctly ignore sockets when comparing
251 a dump (as socket cannot be properly restored anyway). Thanks
252 to Gunther Reiszig <gunther@mit.edu> for reporting the bug.
253
254 4. Fixed restore to correctly access the archive file (-A argument)
255 even when using a remote tape. Thanks to Fabrice Bellet
256 <fabrice@bellet.info> for reporting the bug.
257
258 5. Fixed (again) handling of long (largefile) seeks in rmt.
259 Thanks to Fabrice Bellet <fabrice@bellet.info> for reporting
260 the bug.
261
262 6. Fixed restore corner case when dealing with large block sizes
263 dump is able to write now (-b 1024). Thanks to Fabrice Bellet
264 <fabrice@bellet.info> for reporting the bug.
265
266 7. Fixed a bug preventing dump to access a filesystem having
267 a label exactly 16 bytes in length. Thanks to <minduni@ti-edu.ch>
268 for reporting the bug.
269
270 8. Made dump store dump dates using explicit timezones, fixing a
271 problem with incremental dumps if the timezone is modified
272 between the dumps. Thanks to Philippe Troin <phil@fifi.org> for
273 the bug report and the patch.
274
275 9. Fixed a bug encountered when dumping individual files (not full
276 filesystems or directories) and dangling symbolic links happen
277 to be in the list of files. For as far as dump is concerned,
278 dangling symbolic links are allowed, and are dumped as is.
279 Thanks to Jin-su Ahn <jsahn@ee.snu.ac.kr> for reporting the
280 bug and providing the fix.
281
282 10. Fixed open and creation modes and permissions for QFA and
283 table-of-contents files in dump and restore. Thanks to
284 Philippe Troin <phil@fifi.org> for the patch.
285
286 11. Fixed the archive file descriptor handling enabling it to be 0.
287 This can happen in some cases when shell redirections are used.
288 Thanks to Philippe Troin <phil@fifi.org> for the patch.
289
290 12. Delayed the opening of archive file until after suid had been
291 dropped (fixing a possible security issue if dump is suid).
292 Thanks to Philippe Troin <phil@fifi.org> for the patch.
293
294 13. Fixed the 'S' command handling in the rmt client part.
295 Thanks to Philippe Troin <phil@fifi.org> for the patch.
296
297 14. Modified the end-of-tape script handling to print out statistics
298 (and stop the timer) before launching the eot script. Also, the eot
299 script does not get run anymore when using -M (which makes sense) or
300 when multiple tapes are listed on the command line
301 (-f tape0,tape1,tapen) (which also makes sense).
302 Thanks to Philippe Troin <phil@fifi.org> for the patch.
303
304 15. Relicensed dump/restore under the 'revised' BSD license, as per
305 ftp://ftp.cs.berkeley.edu/ucb/4bsd/README.Impt.License.Change.
306
307 16. Added LZO compression to dump. This new compression method has
308 the advantage of being super fast, thus not killing tape streaming
309 on slow machines. Thanks to Helmut Jarausch
310 <jarausch@igpm.rwth-aachen.de> for the patch and to
311 Markus Oberhumer <markus@oberhumer.com> for giving special permission
312 to include his miniLZO project (GPL licensed) in dump/restore.
313
314 17. Some small buffer overruns fixes in rmt. Thanks to Antonomasia
315 <ant@notatla.demon.co.uk> for reporting the bugs.
316
317 18. Added a special rmt version which can do encryption when writing
318 to tape. Read examples/encrypted_rmt/README for details on
319 how to enable and configure it. Thanks to Ken Lalonde
320 <ken@globalremit.com> for the patch.
321
322 19. Made dump work with 2.5 kernel end of tape early warning semantics.
323 Thanks to Kai Makisara <Kai.Makisara@kolumbus.fi> for the patch.
324
325 20. Fixed a bug which caused dump -w|-W not to work anymore, because
326 the fs_freq and fs_passno fields in /etc/mtab are always set
327 to 0 0. Thanks to Trent Piepho <xyzzy@speakeasy.org> for
328 reporting the bug.
329
330 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
331 =======================================================================
332
333 1. Added a note in the restore man page clarifying the question
334 'set the permissions on the current directory ?' asked by
335 restore at the end of treatment in -i and -x modes.
336
337 2. Fixed the endianess issues when reading compressed tapes.
338 Thanks to Dark Force <daq4th@users.sourceforge.net> for
339 reporting this bug and providing test cases.
340
341 3. Fixed the "ACL won't be dumped" warning message (which showed
342 an extra, unrelated error message). Thanks to Dragan Krnic
343 <dkrnic@lycos.com> for reporting this bug.
344
345 4. Made dump look first into /etc/mtab, then into /etc/fstab
346 when searching for filesystem paths. Also fixed some problems
347 caused by binding mounts or multiple block device mounts.
348 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
349 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
350 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
351 and #160305, Sourceforge bugs #618699 and #679832).
352
353 5. Made dump's -I option accept the value '0' meaning all the
354 read errors will be ignored. This can be useful when running
355 dump from unattended sessions (like cron jobs). Thanks to
356 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
357
358 6. Fixed the output of dump to indicate 'blocks' instead of
359 'tape blocks' in the various messages (blocks are always
360 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
361 made it clearly print the current blocksize at the start of
362 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
363 the suggestions.
364
365 7. Made rmt understand long (largefiles) seeks.
366
367 8. Fixed build with very old versions of libext2fs, where
368 EXT2_FT_* constants were undefined.
369
370 9. Made dump accept the dumpdates path on the command line
371 (-D file option) instead of using only the hardcoded one.
372 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
373 suggestion.
374
375 10. Enabled rmt, LFS, readline, QFA options by default in
376 ./configure. Updated the configure process (new versions
377 of config.guess, config.sub etc).
378
379 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
380 =======================================================================
381
382 1. Changed dump to use fcntl(F_SETLK) style locking instead
383 of flock() when locking the dumpdates file. With the old
384 locking scheme, a local user having read rights on the
385 dumpdates file could be able to do a Denial of Service attack
386 on dump. In order to lock the dumpdates file with the new
387 scheme, the user would need to have write access on the file.
388 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
389 reporting the bug (originally a bugtraq post).
390
391 2. Fixed interactive 'ls' which caused spurious errors warnings
392 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
393 <jorgen_ostling@users.sourceforge.net> for reporting this
394 bug.
395
396 3. Fixed dump's estimate when dealing with sparse inodes.
397
398 4. Modified dump to allow setting a blocksize bigger than 32kB
399 (raised the limit to 1024kB), because newer hardware needs
400 this for proper operation. Thanks to Dirk Traenapp
401 <dtraenapp@users.sourceforge.net> for reporting this.
402
403 5. Fixed a bug causing Dump to stop and report an error if an
404 inode number in the exclude file was followed by some amount
405 of whitespace. Thanks to Jeffrey Sofferin
406 <sofferin@users.sourceforge.net> for reporting this bug.
407
408 6. Fixed a bug which caused restore, in some particular cases,
409 to ask some 'scary' questions and leave a bunch of RSTTMP
410 directories behind when restoring incremental tapes. Thanks
411 to Philippe Troin <phil@fifi.org> for reporting this bug and
412 providing the test cases.
413
414 7. Changed the wording when inodes are excluded from dump:
415 replaced 'Added inode 7 to exclude list' with
416 'Excluding inode 7 (resize inode) from dump', as suggested
417 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
418
419 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
420 ===================================================================
421
422 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
423 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
424 <eros@lamel.bo.cnr.it> for reporting the bug.
425
426 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
427 ===================================================================
428
429 1. Made dump print out the ext2 logical block number in case of a read
430 error, which can be given as an argument to debugfs' ncheck command,
431 in order to find out the affected inode. Added note about this
432 usage in dump's man page.
433
434 2. Fixed a problem in restore when reading tapes written on big
435 endian machines with very old versions of dump. The patch was
436 contributed by George Helffrich <george@geology.bristol.ac.uk>.
437
438 3. Fixed the tape length calculation when using large tapes
439 and compression. Thanks to Georg Lippold
440 <g_lippold@sourceforge.net> for reporting the bug.
441
442 4. Added a new set of examples in dump_on_cd_2 directory, based
443 on dump_on_cd examples but somewhat enhanced, supporting DVD
444 media, and localized in english and german. Thanks to
445 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
446
447 5. Made dump save 32 bit UID/GID. Internally, this was achieved
448 by switching from the old BSD inode format to the new BSD
449 inode format, which means that the tape format was changed.
450 However, since all restore versions out there should
451 transparently support both inode formats, the change should
452 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
453 reporting the bug.
454
455 6. Fixed a lot of warnings in the code shown when compiling
456 with 'gcc -W'. Thanks to Matthias Andree
457 <matthias.andree@stud.uni-dortmund.de> for reporting this.
458
459 7. Fixed a small markup bug in the dump man page. Thanks to
460 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
461 patch.
462
463 8. Rewrote entirely the man pages using the tmac.an macro
464 package (Linux man page format) instead of the original BSD
465 format. They should be now cleaner and easier to modify.
466
467 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
468 ==================================================================
469
470 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
471 non Linux clients were misinterpreted. The description of the problem
472 (incompatible numbering in Linux mtio opcodes) is documented at
473 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
474 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
475 providing an excellent, cross-platform replacement for rmt in his
476 star package.
477
478 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
479 which caused restore to fail when the inode maps were not entirely
480 contained on the first volume (when using really small volumes or
481 when dumping a huge number of inodes).
482
483 3. Fixed a problem in dump, where files in subdirectories of directories
484 that have the nodump flag set (or being excluded with -e/-E) were
485 sometimes incorrectly dumped (instead of being excluded). The original
486 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
487 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
488 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
489 help testing the patch.
490
491 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
492 which makes one able to pipe the output of dump, by the net, to
493 a remote CD-burner server.
494
495 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
496 and use GNU's symbolic syntax over rmt instead of numerical values
497 to assure multiple platform compatibility.
498
499 6. Documented the -d option in restore.
500
501 7. Added a -v (verbose) mode to dump. For now it just prints the number
502 of the inode being dumped, but this could evolve in future versions
503 to include interesting debugging output.
504
505 8. Added a -o flag to restore, which automatically restores the current
506 directory permissions in -i or -x mode, without asking the operator
507 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
508 and Peter Samuel <peters@e-smith.com>.
509
510 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
511 ====================================================================
512
513 1. Fixed a bug in the QFA feature which could have caused
514 a bad tape positionning on restore (causing extra delays in
515 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
516
517 2. Added a small note in the dump man page specifying that
518 there should be no white space between the option letter and
519 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
520
521 3. Made restore work with older versions of the readline library,
522 by checking in configure for several extended symbols. Restore
523 can now be compiled with a readline as old as the 2.0 release
524 (though it may be a good idea to upgrade it to a more recent
525 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
526 reporting the build failures.
527
528 4. Fixed a performance problem with the QFA file creation in
529 dump, which made unnecessary seeks on the tape slowing down
530 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
531 reporting this issue.
532
533 5. Removed the inclusion of some kernel headers in the dump
534 source, which prevented the compile in some kernel/glibc
535 headers/architecture combination. Thanks to Bdale Garbee
536 <bdale@gag.com> for reporting the bug.
537
538 6. Added the appropriate error message when dump fails to
539 open the output file for writing. Thanks to Amith Varghese
540 <amithv@yahoo.com> for reporting this bug.
541
542 7. Made restore able to understand large Solaris ufsdump tapes
543 (containing inodes bigger than 4194304). Sun have introduced
544 an "extension" to the dump tape format when dealing with
545 those inodes, which was uncorrectly handled by Linux restore.
546 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
547 providing a test case.
548
549 8. Added the -m parameter to dump which optimises the output for
550 inodes having been changed but not modified since the last dump
551 ('changed' and 'modified' have the meaning defined in stat(2)).
552 For those inodes, dump will save only the metadata, instead of
553 saving the entire inode contents. Inodes which are either
554 directories or have been modified since the last dump are saved
555 in a regular way. Uses of this flag must be consistent, meaning
556 that either every dump in an incremental dump set have the flag,
557 or no one has it.
558
559 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
560 =======================================================================
561
562 1. Fixed behaviour of dump when exceeding resource limits
563 (SIGXFSZ treatment).
564
565 2. Added the -L flag to restore to allow the user to specify a
566 maximal allowed number of miscompares when using restore
567 with the -C option to check the backup.
568
569 3. Detailed the manual entry for the -N option of restore.
570
571 4. Added the -a flag to restore to make able doing unattended
572 restores in -i or -x mode (automatically walks through the
573 multiple dump volumes).
574
575 5. Extended the QFA mode to work with local files and/or
576 remote tapes and files. This way, restore can know in advance
577 the tape number and the offset for the inodes to extract and
578 can minimize the extraction time by seeking directly to the
579 good tape/offset.
580
581 6. Added the -A <archive> option to both dump and restore,
582 which makes dump to archive a dump table-of-contents in
583 the specified file to be used by restore to determine
584 whether a file is in the dump file that is being restored.
585 (the archive file syntax is also compatible with the
586 Solaris ufsdump generated one).
587
588 7. Small fix in restore making it able to read some (broken ?)
589 Solaris ufsdump tapes.
590
591 8. Fixed dump to correctly recognise the root filesystem when using
592 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
593 for reporting this bug.
594
595 9. Added the -P <file> option to restore to create a
596 Quick File Access file from an already made dump. Patch
597 contributed by Uwe Gohlke <uwe@ugsoft.de>.
598
599 10. Made restore compile and run on Solaris, making it a
600 possible replacement for the standard ufsrestore. Port was
601 contributed by Uwe Gohlke <uwe@ugsoft.de>.
602
603 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
604 =====================================================================
605
606 1. Added a set of backup scripts from Eugenio Diaz
607 <getnito@yahoo.com> in the examples section. It features
608 automatic (cron based) full and incremental dumping of
609 several filesystems on a separate filesystem.
610
611 2. Fixed a off-by-one miscalculation which disabled dumping
612 a one letter subdirectory. Thanks to Chris Siebenmann
613 <cks@utcc.utoronto.ca> for reporting the bug.
614
615 3. Fixed several restore bugs occuring when trying to
616 restore some missing files on the tape. Thanks to Chris
617 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
618
619 4. Fixed --with-ldopts configure argument passing, installing from
620 a separate object directory, makefile cleanups contributed
621 by <splite@purdue.edu>.
622
623 5. Fix a bug which could caused, in some conditions, the highest
624 number inode of a filesystem, to not be dumped. Many thanks
625 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
626 chase this bug.
627
628 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
629 =======================================================================
630
631 1. Added a mini howto from Patrick Walsh in the examples
632 directory.
633
634 2. Minor man pages syntax corrections. Thanks to
635 Chris Danis <screechco@home.com> for reporting the bugs.
636
637 3. Added a script from David B. Peterson <dave@toppledwagon.com>
638 to the examples section. It features dumping several
639 filesystems to a remote tape drive upon ssh.
640
641 4. Added a patch provided by Richard Jones <rich@annexia.org>
642 which allows BRADEMAX (number of read errors tolerated by
643 dump) to be adjusted using the -I option.
644
645 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
646 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
647 patch.
648
649 6. Add the -l option to restore to specify if, when doing a
650 remote restore, the file used to access the backup is a
651 regular file (the defaults being a tape drive). Restore needs
652 to know this information when reading a remote compressed
653 dump. Previously, this information was autodetected, but
654 the autodetection code fails (with ioctl: Inappropriate ioctl
655 for device) when using a non Linux remote box. Thanks to
656 many users and especially to Eros Albertazzi
657 <eros@lamel.bo.cnr.it> for reporting this.
658
659 7. Found a workaroung for the dump deadlock problem (3 childs
660 stuck in pause(), father in read()). The workaround seems
661 to work for me and several beta-testers. If it doesn't work
662 for you, please report back.
663
664 8. Updated the RPM spec file (BuildPrereq, URL etc).
665
666 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
667 ========================================================================
668
669 1. Fixed the permissions of a newly created QFA file by dump.
670
671 2. Cleaned up the source of dump (the external variables
672 definition was a complete mess, making possible to have
673 objects overlap).
674
675 3. Fixed restore to use the full tape volume path when doing
676 a compare (since it changes the working directory to the
677 filesystem being compared in the process).
678
679 4. Added the -q option to dump which makes dump abort
680 whenever it needs operator attention. It should be
681 easier to use dump in scripts with this option.
682
683 5. Detect the use of incompatible options to dump and
684 refuse them (like -a and -B options together).
685
686 6. Added bzip2 compression to dump/restore (use option -j level
687 to select it). Note that you will need a restore version
688 >= 0.4b24 in order to restore a bzip2 compressed dump.
689 The same warning as for the zlib compression applies:
690 the tape format of a bzip2 dump is not compatible with the
691 original BSD tape format.
692
693 7. Fixed a overflow problem in dump corrupting the dump when
694 very large files were encountered. Thanks to Vladimir Nadvornik
695 for the bug report.
696
697 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
698 kernel buffer/page cache before starting the dump, helping
699 a bit those who use dump on mounted filesystems. Thanks to
700 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
701 for this suggestion.
702
703 9. Updated the RPM spec file following the RedHat changes
704 (dynamically linked binaries now in /usr/sbin etc).
705
706 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
707 which enables restore to recognise multi volume compressed dumps
708 done on CD. Included his scripts for dump (which pipe the dump
709 output directly into cdrecord) and restore. There is now
710 possible to to backups to CD on the fly!
711
712 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
713 ===================================================================
714
715 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
716 Marcus Meissner (Caldera International Security Dept.).
717
718 2. Implement the Sun rmt extended protocol. Patch submitted
719 by Ian Gordon <iangordon@users.sourceforge.net>.
720
721 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
722 the linux kernel header. This ensures that dump always has the
723 latest version of this file. Patch submitted by
724 Andreas Dilger <adilger@turbolinux.com>.
725
726 4. Report any filesystem present in either /etc/fstab with a
727 positive passno or /etc/dumpdates in dump -w output.
728 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
729
730 5. Fixed the looping problem in dump introduced in the
731 previous version.
732
733 6. Changed the -B option of dump to limit the size of
734 _compressed_ output per volume if compression is on.
735 Patch contributed by Helmut Jarausch
736 <jarausch@igpm.rwth-aachen.de>. Note however that, since
737 it is impossible to predict the size of the compressed
738 data before doing the compression, the -B limit is a bit
739 conservative.
740
741 7. Fixed a bug in reading the operator typed file/tape path for
742 the next volume in restore.
743
744 8. Implemented a "-F script" option to restore which permits the
745 user to specify a script which will be launched at the
746 beginning of each tape, useful for automatic programming of
747 tape changers for example. See the restore man page for the
748 script parameters and return codes.
749
750 9. Small fix for the QFA routines provided by Uwe Gohlke
751 <uwe@ugsoft.de>, and some recommendations for QFA uses in
752 the man pages.
753
754 10. Fixed the multivolume restoring where making a mistake
755 to the 'Mount next tape' prompt caused several blocks to
756 be lost.
757
758 11. Enhanced the -e option of dump to take as a parameter a
759 comma separated list of inode numbers.
760
761 12. Added the -E option to dump which specify a file containing
762 inode numbers to exclude from the dump.
763
764 13. Fixed the compressed multi-volume dump + restore.
765
766 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
767 ==================================================================
768
769 1. Made dump report the number of blocks dumped per volume.
770 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
771
772 2. Fix a bug in dump where waiting too much at the 'change volumes'
773 question caused the volume to be rewritten. Thanks to
774 Graham Reed <greed@users.sourceforge.net> for reporting the
775 bug and providing a patch.
776
777 3. Added a compression option to dump, contributed by Jerry
778 Peters <gapeters@worldnet.att.net>.
779
780 WARNING: the compression works only on files/pipes and
781 drives supporting variable block size.
782
783 WARNING2: tapes created using the compression feature are
784 incompatible with the standard dump tape format, and a
785 version of restore >= 0.4b22 will be needed for extraction.
786
787 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
788 architectures. Thanks to Paul Slootman <paul@debian.org> for
789 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
790 it upstream.
791
792 5. Many cleanups (CPP defines, const char warnings, check of
793 ext2fs COMPAT flags, time_t cleanups, added libext2 version
794 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
795
796 6. Made --prefix option work in configure. All the install path
797 are now based on the configure parameters.
798
799 7. Added the Quick File Access mode in dump/restore, contributed
800 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
801 a file tape position for each inode, and this file is used by
802 restore (if called with parameter Q and the filename)
803 to directly position the tape at the file restore is currently
804 working on. This saves hours when restoring single files from
805 large backups, saves the tapes and the drive's head. Use
806 --enable-qfa option of configure to compile in the QFA support.
807
808 8. Added the possibility to dump several files and directories
809 in a single invocation of dump. Thanks to Uwe Gohlke
810 <uwe@ugsoft.de> for implementing this option.
811
812 9. Fixed the dumping and restoring of files with holes
813 (files ending with holes didn't get dumped or restored
814 correctly).
815
816 10. Fixed a socket descriptor leak in dump, which leaved opened
817 3 file descriptors per dump process (and there is one dump
818 process per tape).
819
820 11. Fixed dump large file system support, by explicit use of
821 open64/lseek64/etc functions (explicit use needed because
822 e2fsprogs libraries don't behave well when compiled with
823 FILE_OFFSET_BITS=64).
824
825 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
826 ======================================================================
827
828 1. Fixed some bugs in the dump -w|-W logic introduced by
829 the previous version. Thanks to Andreas Dilger
830 <adilger@turbolinux.com> for his help on this one.
831
832 2. Fixed again a compilation problem when using old e2fs
833 headers (filesystem label related). Thanks to many users
834 who reported this stupid error.
835
836 3. Fixed a build problem on old lib5 based systems dealing with
837 _PATH_MNTTAB being undefined. Thanks to John Adams
838 <johna@onevista.com> for reporting the bug.
839
840 4. Improved the error detection in restore when creating
841 the temporary files in TMPDIR. Now restore will corectly
842 report a 'no space left on device' error instead of
843 strange errors which could imply an archive corruption.
844 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
845 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
846
847 5. Added the throughput information in dump's progression
848 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
849 for the patch.
850
851 6. Use libext2fs's inode scanning routines, which are particularly
852 robust when dealing with errors (especially when having some
853 bad blocks on the filesystem). Many thanks to Colin
854 <colin@horizon.com> for reporting the bug, investigating
855 the issues, finding a workaround, writing the patch and
856 fully testing it... (of course, if this patch will break
857 anything, he is to blame for :-)).
858
859 7. Made dump and restore LFS aware. Dump can dump a filesystem
860 containing large files, generate a large file on output and
861 restore can restore them. This feature is not enabled by
862 default, use --enable-largefile option of configure to enable
863 it (you need to have a LFS aware glibc though). Thanks to
864 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
865 and to Theodore T'so <tytso@valinux.com> for his always
866 useful thoughts.
867
868 8. Made dump ask upon a tape write error if it should rewrite
869 the current volume (assume this is a bad tape) or if it should
870 assume an end-of-tape condition (useful for tape drives which
871 behaves badly at the end of the tape). Thanks to Andreas
872 Dilger <adilger@turbolinux.com> for the suggestion.
873
874 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
875 =======================================================================
876
877 1. Fixed a small compilation problem due to a change
878 in the definintion of the struct sigaction in
879 glibc 2.0 / libc5. Thanks to Gunther Schlegel
880 <schlegel@riege.de> for reporting the bug and to
881 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
882 a fix.
883
884 2. Modified the label and uuid analysis in order to be
885 self-contained (independant of kernel/libc headers). This
886 should fix the compile with older kernel/libc headers and
887 will preserve the functionality. Thanks to Bernhard Erdmann
888 <bernhard.erdmann@gmx.de> for reporting the bug.
889
890 3. The 'exclude inode' option, if applied to a directory
891 inode, excludes now correctly all the leaves of this
892 directory. Thanks to John R. Dennison
893 <gerdesas@users.sourceforge.net> for reporting the bug.
894
895 4. Fixed the '-e' option to disable the possibility
896 to exclude the root inode (which causes the dumps to
897 be unreadable by restore). Prevented array overflow
898 when multiple -e options are used.
899
900 5. Fixed dump to correctly interpret a filesystem argument
901 which is a mountpoint and it is not an absolute path
902 (as specified in the fstab). Thanks to Bernhard R. Erdmann
903 <be@berdmann.de> for reporting the bug.
904
905 6. Made dump able to backup files larger than 2 GB. Note that
906 dump still doesn't cope with files larger than 4 GB.
907
908 7. Restore the real uid and gid before invoking an external
909 RSH program (big hole when dump or restore is suid root!).
910
911 8. Get the values for the default tape device and the location
912 of fstab file from the system headers. Thanks to
913 Andreas Dilger <adilger@turbolinux.com> for the patch.
914
915 9. Made dump -w|-W report all recognized filesystems
916 present in either /etc/fstab or /etc/dumpdates, and present
917 the list in the same order as in fstab file. Thanks
918 to Andreas Dilger <adilger@turbolinux.com> for the patch.
919
920 10. Made dump's -a (automatic end of tape detection) the
921 default. Specifying one of -B, -c, -d or -s options will
922 override the EOM detection. Thanks to Andreas Dilger
923 <adilger@turbolinux.com> for the patch.
924
925 11. Save the ext2 filesystem label into the dump volume label.
926 Specifying a volume label on the command line (-L option)
927 will override this feature. Thanks to Andreas Dilger
928 <adilger@turbolinux.com> for the patch.
929
930 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
931 =====================================================================
932
933 1. Fixed the signal handling in dump (which I broke in 0.4b17)
934 which was causing several strange dump failures (dump
935 hanged or segmentation faults etc).
936
937 2. Specified the default blocksize in the dump man page.
938
939 3. Changed two info messages of restore to be written on stdout
940 instead of stderr in order to leave stderr only for errors
941 or warnings. Thanks to Stephen Atwell
942 <satwell@urbana.css.mot.com> for the suggestion.
943
944 4. Corrected an off by one calculation which prevented
945 dumping a directory having a 1 character name length.
946 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
947 for reporting the bug.
948
949 5. Reinforce dump in order to allow only level 0 and no
950 -u option when dumping a subdirectory, and document
951 this in the man page. Thanks to Bernhard Erdmann
952 <bernhard.erdmann@gmx.de> for reporting the bug.
953
954 6. Fixed a small harmless bug in restore which caused
955 in some conditions hard links to be restored several
956 times (therefore generation some warning messages).
957 Thanks to Kenneth Porter <shiva@well.com> for
958 reporting the bug.
959
960 7. Updated the RPM spec file to the latest RedHat version,
961 providing FHS packaging and other cosmetic changes.
962 You will need RPM version >= 3.0.5 in order to build the RPM.
963
964 8. Updated the configure script to check for libtermcap
965 before checking for libreadline, because we need this
966 library in order to compile the readline support.
967
968 9. Made dump understand the LABEL= and UUID= notation
969 both in /etc/fstab and on the command line. Note that
970 you will need the /proc filesystem in order to use
971 these notations. Thanks to Erik Troan <ewt@redhat.com>
972 for providing the patch.
973
974 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
975 ===================================================================
976
977 1. Fixed a potential buffer overflow in restore. Thanks
978 to Stan Bubrouski <satan@fastdial.net> for reporting
979 the bug.
980
981 2. Fixed a readline-related bug which prevented
982 'cat DUMPFILE | restore -i -f -' from working. Thanks
983 to Charles Karney <karney@users.sourceforge.net>
984 for the bug report.
985
986 3. Changed a few "panic" into "exit", causing restore to
987 be more stable against some attacks (like the last one
988 reported on Bugtraq, although the last version of restore
989 was not vulnerable - just dumped core). Thanks to
990 Andreas Hasenack <andreas@conectiva.com.br> for reporting
991 the bugs to me.
992
993 4. Removed the suid-root bit on dump and restore in the
994 default build (and generated RPMs). It should be safer
995 now. Users who need the suid-root capabilities in order
996 to make network backups should read first the man page
997 of dump and enable the suid bit manually.
998
999 5. Added -ltermcap to the compile parameters for restore
1000 when using readline support, in order to make the compile
1001 process work on some systems (SuSE ?). Thanks to
1002 Patrik Schindler <poc@pocnet.net> for reporting the bug.
1003
1004 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
1005 ====================================================================
1006
1007 1. The -F script is called now *only* at the end of a tape,
1008 not at the end of the dump. Thanks to Christian Haul
1009 <haul@informatik.tu-darmstadt.de> for the bug report.
1010
1011 Normally, the device name and the current volume number
1012 are passed on the command line of the script. However,
1013 if you want the old style script (with no arguments
1014 passed on the command line), you can enable it in
1015 configure with the --enable-oldstylefscript.
1016
1017 2. Use posix signal handling to preserve dump functionality
1018 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
1019 the patch.
1020
1021 3. Made the exit code of restore in case of a 'C'ompare
1022 command reflect the result of the compare. An exit status
1023 of 0 means the dump archive is correct, 1 means tape errors,
1024 2 means that some files were modified. Thanks to Kenneth Porter
1025 <shiva@well.com> for the suggestion.
1026
1027 4. Made (finally) quotes work in interactive restore.
1028
1029 5. Small fixes in order to allow dump to compile with a
1030 really old e2fsprogs version. Thanks to Ian Zimmerman
1031 <itz@speakeasy.org> for the bug report.
1032
1033 6. Add GNU readline capabilities to interactive restore.
1034 Use configure's --enable-readline flag to enable this feature.
1035 Thanks to Patrik Schindler <poc@pocnet.net> for the
1036 suggestion.
1037
1038 7. Do the compare on the fly in restore 'C' mode (this will
1039 allow not to exhaust the available /tmp space when
1040 ccmparing large files). Thanks to Kenneth Porter
1041 <shiva@well.com> for the suggestion.
1042
1043 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
1044 ====================================================================
1045
1046 1. Fixed some several duplicate 'DUMP: DUMP:' in the
1047 output of dump.
1048
1049 2. Corrected the estimation of blocks to dump. Note that
1050 this is still wrong for large directory inodes, due
1051 to the size of a BSD directory entry on the tape
1052 (estimation is based on the size of the ext2 directory,
1053 which is smaller because it doesn't contain the null
1054 character at the end).
1055
1056 3. Made dump report the total number of MB written to
1057 tape at the end of run. Thanks to W. Reilly Cooley
1058 <wcooley@nakedape.cc> for the patch.
1059
1060 4. Added the -X option to restore giving the possibility
1061 to read the names of the files to be extracted/listed
1062 from a text file (in addition of the command line).
1063 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
1064 patch.
1065
1066 5. Added the device name and the current volume number
1067 as arguments to the end of tape script (dump -F option).
1068
1069 6. Made the multi-volume dump work again (after having
1070 broken it in 0.4b15).
1071
1072 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
1073 ===================================================================
1074
1075 1. Added a prompt command in interactive restore mode. Thanks
1076 to Andreas Dilger <adilger@home.com> for the patch.
1077
1078 2. Fixed a buffer overflow problem in dump (caused by
1079 not checking the size of the filesystem parameter).
1080 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
1081 reporting this on Bugtraq (and to several dump users
1082 who forwarded me his mail).
1083
1084 3. Added the '-F script' option to dump in order to
1085 launch a script at the end of each tape (to be used
1086 with a tape changer, or to notify the sysadmin by
1087 pager etc.).
1088
1089 4. Fixed a bug in restore compare code caused by the changes
1090 I made in 0.4b14.
1091
1092 5. Fixed the treatment of options using the old BSD syntax
1093 in both dump and restore.
1094
1095 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
1096 =======================================================================
1097
1098 1. Fixed a bug in dump which may have caused invalid deleted
1099 directories to be dumped out if they were deleted after the
1100 mapping pass. This could occure on active filesystem and lead
1101 to heap corruption (causing dump malfunction in many possible ways).
1102 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
1103 bug and submitting the patch.
1104
1105 2. Fixed the handling of the filesystem parameter in dump. A
1106 '/mnt/disk' parameter caused the disk contents to be dumped,
1107 but a '/mnt/disk/' parameter caused the mountpoint directory
1108 to be dumped (generally an empty directory).
1109
1110 3. Improved the output of dump in order to tell which directory
1111 it is currently dumping (when dumping a subtree).
1112
1113 4. Added the '-e' exclude inode option to dump. Thanks to
1114 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
1115
1116 5. Added a REPORTING-BUGS file in order to provide a guide
1117 on how to correctly report a bug in dump/restore.
1118
1119 6. Fixed a restore problem when restoring a hard link to an inode
1120 having the immutable or append-only attribute set. Thanks to
1121 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
1122
1123 7. Fixed a compatibility problem between dumps made on little
1124 endian machines (the format was correct) and big endian
1125 machines (the format was incorrect). This fix break the
1126 compatibility with the older dumps made on big endian
1127 machines (sparc, m86k, ppc etc). For the first time in
1128 linux dump's history (I believe), the dumps made by *BSD,
1129 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
1130 so interchangeable. Thanks to Rob Cermak
1131 <cermak@ahab.rutgers.edu> for submitting the bug and
1132 helping me test the fix.
1133
1134 8. Fixed the way dump reports the remaining percent/time, if
1135 the number of blocks actually dumped exceeds the estimated
1136 number of blocks. Thanks to Jean-Paul van der Jagt
1137 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
1138
1139 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
1140 ======================================================================
1141
1142 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
1143 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
1144
1145 2. Removed the requirement to build the RPM as root from the
1146 spec file. Thanks to Christian Weisgerber
1147 <naddy@mips.rhein-neckar.de> for submitting this.
1148
1149 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
1150 causing dump to block when dumping really huge filesystems.
1151 Many thanks to Patrik Schindler <poc@pocnet.net> for
1152 helping me find this bug.
1153
1154 4. Fixed the treatment for an interrupt signal when dump access
1155 the remote tape through RSH. Thanks to Christian Weisgerber
1156 <naddy@mips.rhein-neckar.de> for providing the patch.
1157
1158 5. Fixed a bug which was causing dump/restore to display
1159 garbage characters instead of the remote host name.
1160
1161 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
1162 =====================================================================
1163
1164 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
1165 <kukuk@suse.de> for submitting the patch.
1166
1167 2. Fix for the exit code when using the size estimate option of
1168 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
1169 submitting the patch.
1170
1171 3. Handle EINTR in atomical reads/writes in dump, which was causing
1172 dump to fail on some systems. Thanks to Eric Jergensen
1173 <eric@dvns.com> for reporting the bug and submitting the patch.
1174
1175 4. Handle more than 16 characters for the device names in dumpdates.
1176 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
1177 tracking down the problem and proposing the solution.
1178
1179 5. Fixed a bug in dump which prevented the creation of the
1180 dumpdates file when doing a 0-level dump without already
1181 having a dumpdates file. Thanks to Patrik Schindler
1182 <poc@pocnet.net> for reporting the bug.
1183
1184 6. Changed the way dump 'S' flag reports the size estimate
1185 from number of blocks into bytes (making it compatible
1186 with the Solaris version, and simplifying things for
1187 amanda users). Thanks to Jason L Tibbitts III
1188 <tibbs@math.uh.edu> for reporting the bug.
1189
1190 7. Fixed a compatibility problem in linux/alpha dump tape format.
1191 Now the linux/alpha dump are (again) compatible with the
1192 other dump formats. But this breaks compatibility with
1193 older dumps made on alpha. Thanks to Mike Tibor
1194 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
1195
1196 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
1197 ======================================================================
1198
1199 1. Added a '--enable-kerberos' to configure.
1200
1201 2. Added a 'S' option to dump which determines the amount of space
1202 that is needed to perform the dump without actually doing it, similar
1203 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
1204 <cermak@ahab.rutgers.edu>.
1205
1206 3. Added a 'M' multi-volume option to dump and restore which enables
1207 dumping to multiple files (useful when dumping to an ext2
1208 partition to make several dump files in order to bypass the 2GB
1209 file size limitation). The 'f' argument is treated as a prefix and
1210 the output files will be named <prefix>001, <prefix>002 etc. With
1211 the 'M' flag, restore automatically selects the right file without
1212 asking to enter a new tape each time.
1213
1214 4. Fixed a memory leak which was causing dump to grow very big
1215 (270MB when dumping a 10GB filesystem...). Thanks to Jason
1216 Fearon <jasonf@netrider.org.au> for reporting the bug.
1217
1218 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
1219 ======================================================================
1220
1221 1. Make configure test if the system glob routines support
1222 extended syntax (ALTDIRFUNC). If not, use the internal glob
1223 routines instead of system ones. Thanks to Bernhard Sadlowski
1224 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
1225 and helping me resolve this and other minor libc5 compiling
1226 glitches.
1227
1228 2. Fix a problem when dumping a ext2fs with the 'filetype'
1229 feature enabled. Thanks to Patrick J. LoPresti
1230 <patl@cag.lcs.mit.edu> for reporting the bug and to
1231 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
1232
1233 3. Made the nodump flag work on directories. A directory which
1234 has the nodump flag gets never dumped, regardless of its
1235 contents.
1236
1237 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
1238 which allows dump on an active ext3 filesystem. However, this
1239 is a "quick and dirty" patch which enables backup of an ext3
1240 filesystem through the ext2 compatibility (by ignoring the
1241 NEEDS_RECOVERY bit). The journal file is not recognized and
1242 it is dumped (it should not).
1243
1244 5. Test the superblock compatibility flags when dumping, in order
1245 to be sure that we know how to deal with specific features.
1246
1247 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
1248 ====================================================================
1249
1250 1. Use lchown instead of chown, fixing a possible security problem
1251 when restoring symlinks (a malicious user could use this
1252 to deliberately corrupt the ownership of important system files).
1253 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
1254 this and providing the patch.
1255
1256 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
1257 ====================================================================
1258
1259 1. Put dump sources under CVS, added Id tags in all files so
1260 one can use 'ident' on binary files.
1261
1262 2. Added the dump/restore version in the usage text so one can
1263 easily verify the version he is using.
1264
1265 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
1266 a va_start/va_end problem on linux-ppc (always call va_start
1267 va_end in pairs each time we use a vararg function).
1268
1269 4. Added again the DT_* constants because old libc does not
1270 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
1271 for submitting the bug report.
1272
1273 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
1274 this should enable dumping big (huge) filesystems.
1275
1276 6. Added the RSH environment variable in order to be able to
1277 use a rsh replacement like ssh when doing remote backups (and
1278 bypass the security limitations of rcmd). Now you can do remote
1279 backups without being root (or making dump setuid root).
1280
1281 7. Modified again the way dumpdates works. For incremental dumps,
1282 we need to read dumpdates even if we are not using 'u' option.
1283 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
1284 this should work.
1285
1286 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
1287 ===================================================================
1288
1289 1. Removed the 'k' flag from the restore 'about' text if kerberos
1290 was not compiled in.
1291
1292 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
1293 to them (fsetflags takes a char*, setflags an open fd!).
1294
1295 3. (f)setflags is called only if the flags aren't empty. If the
1296 file is a special file, a warning is printed, because changing
1297 flags implies opening the device. Normally, a special file
1298 should not have any flag... (Debian bug #29775, patch provided
1299 by Abhijit Dasgupta <abhijit@ans.net>).
1300
1301 4. Made possible to dump a file system not mentioned in /etc/fstab.
1302 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
1303
1304 5. Changed the default behaviour to not create dumpdates
1305 unless 'u' option is specified. Removed the old "debian-patch"
1306 which provided the same thing. (Debian bug #38136, #33818).
1307
1308 6. Removed all those dump*announce, since they were getting old...
1309
1310 7. Added warning messages if dumpdates does not exist and
1311 when an inferior level dump does not exist (except for a level 0
1312 dump).
1313
1314 8. Debugged the glob calls in interactive mode: restore used a
1315 dirent struct which was different from the /usr/include/dirent.h
1316 one (this used to work, is it a glibc2 change?), so none of the
1317 compat glob (which used /usr/include/dirent.h) or the system glob
1318 worked. Restore use now the system dirent (and the system
1319 DT_* constants), which are compatible with BSD ones.
1320
1321 9. Added a configure flag (--with-dumpdatespath) to specify
1322 the location of dumpdates file. By default, it is
1323 /etc/dumpdates.
1324
1325 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1326 included the current date/version in man pages.
1327
1328 11. Corrected the estimation of remaining time when
1329 the operator doesn't change the tapes quickly enough. This
1330 was an old bug, I thought I corrected it, and discovered
1331 that in fact it was corrected in two different places, so
1332 the results canceled each other...
1333
1334 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1335 ===================================================================
1336
1337 1. Integrated multiple patches from RedHat, Debian and SuSE:
1338
1339 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1340 isn't specified.
1341 - several fixes in the man pages.
1342 - update the default tape device to /dev/st0.
1343 - many updates for Linux Alpha (byte ordering, size_t etc).
1344 - buffer overruns.
1345 - use environment variable for TMPDIR (instead of /tmp).
1346 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1347 - workaround egcs bug (RedHat bugs #4281 and #2989).
1348 - wire $(OPT) throughout Makefile's.
1349
1350 2. Upgrade the dump revision to 1, making possible to dump filesystems
1351 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1352
1353 3. Fix some compile warnings, prototype all functions.
1354
1355 4. Use glibc err/glob instead of internal compatibility
1356 routines (only if available).
1357
1358 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1359 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1360
1361 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1362 ======================================================================
1363
1364 1. Integrated the changes from FreeBSD-3.1-RELEASE
1365 (mostly bug fixes, buffer overruns, dump has now an "automatic
1366 tape length calculation" flag, dump/restore can use kerberos now
1367 (this is NOT tested), use environment variables for TAPE and
1368 RMT etc.).
1369
1370 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1371
1372 3. Corrected a bug in restore when using 'C' option with multi-volumes
1373 tapes (files splited accros two tapes give "size changed" errors
1374 when comparing).
1375
1376 4. Corrected the long standing bug when dumping multiple tapes.
1377 This works for me, needs further testing.
1378
1379 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1380 ====================================================================
1381
1382 1. Dump now runs correctly on kernels 2.1.x
1383 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1384
1385 Changes between versions 0.4b2 and 0.4b3
1386 ========================================
1387
1388 1. Use realpath() if available
1389
1390 2. Report statistics
1391
1392 Changes between versions 0.4b1 and 0.4b2
1393 ========================================
1394
1395 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1396 the patch)
1397
1398 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1399
1400 3. Fixed configure.in to correctly handle --enable-rmt
1401
1402 Changes between versions 0.3 and 0.4b1
1403 ======================================
1404
1405 1. Integrated the changes from 4.4BSD-Lite2
1406
1407 2. Integrated the patches from Debian and Red Hat
1408
1409 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1410
1411 4. Changed dump to use the Ext2fs library to get block addresses. This
1412 should solve the endianness problem on SparcLinux.
1413
1414 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1415 distribution's one) to use autoconf
1416
1417 6. Fixed a few minor bugs
1418
1419 Changes between versions 0.2e and 0.2f
1420 ======================================
1421
1422 1. Added the creation of named pipes (fifos) in restore.
1423
1424 2. Added the -N flag in the restore manual page.
1425
1426 3. Added the file kernel.patch which contains the llseek() optimization
1427 patch for 1.2.x kernels.
1428
1429 4. Fixed a bug in the restoration of symbolic links: owner and group were
1430 not restored.
1431
1432 5. Integrated some changes from FreeBSD 2.2.
1433
1434 6. Added a call to ftruncate() after restoring each file to restore
1435 correctly files ending by a hole.
1436
1437 Changes between versions 0.2d and 0.2e
1438 ======================================
1439
1440 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1441 with owner = root (0) and group = root/wheel/whatever (0).
1442
1443 Changes between versions 0.2c and 0.2d
1444 ======================================
1445
1446 1. Dump is now able to backup 2GB+ filesystems.
1447
1448 2. Dump and restore can now be linked as static binaries.
1449
1450 Changes between versions 0.2b and 0.2c
1451 ======================================
1452
1453 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1454
1455 Changes between versions 0.2a and 0.2b
1456 ======================================
1457
1458 1. Really fixed the bug that I should have corrected in 0.2a.
1459
1460 2. Enabled optimization again.
1461
1462 Changes between versions 0.2 and 0.2a
1463 =====================================
1464
1465 1. Disabled the optimization during compilation.
1466
1467 Changes between versions 0.1 and 0.2
1468 ====================================
1469
1470 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1471 the fs_type field (actually, I modified the file fstab.c to make it
1472 use the mntent functions).
1473
1474 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1475 made by dump should be readable by the BSD restore and backups made
1476 by the BSD dump should be readable by restore. Unfortunately, this
1477 means that the dump format has changed between version 0.1 and version
1478 0.2 :-(
1479
1480 3. Dump is now able to backup a subtree, it is no longer limited to whole
1481 filesystems like the BSD version.
1482
1483 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1484 than 2 GB.
1485
1486 Changes between versions 0.0 and 0.1
1487 ====================================
1488
1489 1. Now create links rdump and rrestore during the `make install' step.
1490
1491 2. Linux port specific bugs added to the manual pages
1492
1493 3. Incorrect estimation of the number of tapes blocks fixed when doing
1494 incremental backups.
1495
1496 4. Better ls-like format in restore in interactive mode.