Fix code to not require ISO C99.
[dump.git] / CHANGES
1 $Id: CHANGES,v 1.275 2005/05/02 15:13:40 stelian Exp $
2
3 Changes between versions 0.4b39 and 0.4b40 (released May 2, 2005)
4 =================================================================
5
6 1. Changed restore to emit warnings (instead of emitting a fatal
7 error) if a file (or a directory) is unavailable for a
8 comparision (if the user doesn't have the necessary permissions
9 to access it for example). Thanks to Kenneth Porter
10 <shiva@sewingwitch.com> for the bug report.
11
12 2. Re-done the 'do not save directory entries to non-dumped inodes
13 (excluded from dump)' feature. The previous implementation
14 worked well for excluded directories but not for regular files.
15 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
16 report.
17
18 3. Fixed a bug in dump where the tape size was miscalculated when
19 the user used -d/-s to specify the tape characteristics. Thanks
20 to Philip Goisman <goisman@physics.arizona.edu> for reporting
21 the bug.
22
23 4. Fixed another bug introduced in restore with the hashtree
24 implementation. This one caused restore to stop saying
25 "removenode: non-empty directory" in some cases.
26
27 5. Added support for dumping and restoring ext2/3 extended
28 attributes (EA), like the access control lists (ACL) or
29 the security labels used by SELinux.
30
31 Changes between versions 0.4b38 and 0.4b39 (released January 21, 2005)
32 ======================================================================
33
34 1. The newly added dump_on_cd_3 example was buggy, replace it
35 with an updated version from Andrew Basterfield
36 <bob@cemetery.homeunix.org>.
37
38 2. Made restore to chdir() back into the initial directory when
39 dumping core while aborting a comparision operation. The
40 previous behaviour was to write the corefile at the root of
41 the directory being compared, which could very well be
42 read only and preventing the corefile generation. Thanks
43 to Kenneth Porter <shiva@sewingwitch.com> for the bug report.
44
45 3. Silenced the failure to call fgetflags() when comparing an
46 entry which has no ext2 attributes (as in lsattr()).
47
48 4. Fix a brown paper bug in restore -C which broke restore and
49 caused modifications on the filesystem being compared
50 (directories containing a file with the same name as the
51 directory get renamed to RSTTMP...). Thanks to Kenneth Porter
52 <shiva@sewingwitch.com> for finding the bug and helping me
53 reproduce it.
54
55 5. Made restore -C force the -N flag (no writing allowed on
56 the disk) in order to prevent more bugs like the above one.
57
58 Changes between versions 0.4b37 and 0.4b38 (released January 7, 2005)
59 =====================================================================
60
61 1. Fix a couple of troff syntax bugs in the man pages.
62 Thanks to Eric Raymond <esr@thyrsus.com> for the patch.
63
64 2. Made restore use either libncurses or libtermcap, depending
65 on which one is available at configure time.
66
67 3. Fixed restore negative size display bug when comparing a
68 dump containing files over 2GB. Thanks to Steve Bonds
69 <sbonds@users.sourceforge.net> for the bug report.
70
71 4. Do not save directory entries to non-dumped inodes
72 (excluded from dump). This will eliminate the 'missing
73 file' warnings when doing 'restore -C'.
74
75 5. Fix dump crash when backuping a huge (2TB) filesystem,
76 due to a bogus calculation on the inode map size.
77 Thanks to Kevin B. Haines <K.B.Haines@rl.ac.uk> for
78 submitting the bug and testing the fix.
79
80 6. Fix a problem in restore where the final \0 in the symbolic
81 link names could have been lost, generating corrupt filenames.
82 Thanks to Kyle Wilson <kyle.wilson@amd.com> for reporting the
83 bug.
84
85 7. Implemented a hash list for the directory names in restore.
86 The linear list used before caused problems in interactive
87 restores when dealing with directories having thousands of
88 entries. Thanks to Brian Ristuccia <bristuccia@starentnetworks.com>
89 for reporting the bug.
90
91 8. Improved restore -C, this time including the directory
92 attributes into the comparision.
93
94 9. Made restore understand tapes containing EA/ACLs (which will
95 be dumped by the next version of dump). In this version
96 extended attributes on the tape are ignored, for full EA/ACL
97 support wait for the next version or try the experimental EA
98 patch.
99
100 Changes between versions 0.4b36 and 0.4b37 (released July 7, 2004)
101 ==================================================================
102
103 1. Added the --enable-staticz configure option which enables
104 dump and restore to be linked with static versions of
105 libz and libbz2 (and dynamic versions of all other libraries).
106 This will make Debian users happy, because libz and libbz2
107 were the only needed libraries living in /usr, all the
108 others live in /lib. In case of system emergency, it is
109 better not to have to rely on an extra filesystem.
110 Thanks to Bdale Garbee <bdale@gag.com> for the suggestion.
111
112 2. Fix compilation on (at least the Linux Debian port to) AMD64.
113 (<ext2fs/ext2_types.h> defines some types (__s64 and __u64)
114 that are also defined by <linux/types.h> (<asm/types.h>) and
115 they conflict).
116
117 3. Make dump's reading of the dumpdates file a bit more robust,
118 preventing dump from crashing when the dumpdates file has
119 been modified by hand.
120
121 4. Fixed some offset calculations in dump code which could
122 lead to "bread lseek errors" on large filesystems. Thanks
123 to Bruce Lowekamp <lowekamp@users.sourceforge.net> for
124 reporting this bug and debugging the issue.
125
126 5. Made dump use the blkid library when searching for devices
127 by label or uuid instead of dump's own routines.
128
129 6. Corrected a bug in dump where a wrong LABEL=... line in
130 /etc/fstab could prevent dump from dumping unrelated
131 filesystems. Thanks to Bruce Lowekamp
132 <lowekamp@users.sourceforge.net> for reporting the bug.
133
134 Changes between versions 0.4b35 and 0.4b36 (released April 21, 2004)
135 ====================================================================
136
137 1. Fixed dump compilation with old gcc versions. Thanks to
138 Mike Castle <dalgoda@users.sourceforge.net> for the patch.
139
140 2. Fixed some warnings (howmany, roundup, powerof2 redefined)
141 when compiling against a recent glibc version.
142
143 3. Fixed a bug in restore preventing the read of a dump tape
144 written with Solaris 7 ufsdump. Thanks to Patrick Higgins
145 <phiggins@transzap.com> for reporting the bug and providing
146 the test case.
147
148 4. Changed dump to enable the creation of volumes of different
149 sizes in a single run (make -B accept a list of values).
150 Patch contributed by Florian Zumbiehl <florz@gmx.de>.
151
152 5. Use the glibc provided minor() and major() macros instead
153 of our own bitmask implementation. This should be safe for
154 when the major/minor namespace will migrate to 32 bits.
155 Thanks to Zhang Jun <zhangjun@nanjing-fnst.com> for reporting
156 the bug.
157
158 6. Made explicit in the dump man page that dump will not create
159 a remote file, it will only write to an already existing one.
160
161 7. Another try at making size estimates better again.
162
163 8. Put back the inconditional running of the end-of-tape
164 script which was changed in 0.434 to be run only when -M
165 or multiple -f were NOT specified. Some users rely on this
166 feature even when it is combined with -M/-f.
167
168 9. Fixed restore when restoring huge backups (where rstdir...
169 temporary files are over 2GB). Thanks to
170 Raphael Jaffey <rjaffey@artic.edu> for reporting this,
171 debugging the issue and making the patch.
172
173 10. Made restore understand FreeBSD UFS2 tapes. Thanks to
174 David <vrtin@users.sourceforge.net> for submitting the bug
175 and providing a test case.
176
177 11. Made dump work with an arbitrary integer level (no more 10
178 levels only limitation). Thanks to Matthew
179 <msvincen@midway.uchicago.edu> for the patch.
180
181 Changes between versions 0.4b34 and 0.4b35 (released December 21, 2003)
182 =======================================================================
183
184 1. Added a note in the dump man page saying that the default
185 blocksize can be 32 if -d is used with a high density argument.
186 Thanks to Antonios Christofides <A.Christofides@itia.ntua.gr>
187 for the patch.
188
189 2. Fixed configure to correctly understand CPPFLAGS, CFLAGS,
190 LDFLAGS environment variables. Thanks to Arcady Genkin
191 <antipode@users.sourceforge.net> for reporting the bug.
192
193 3. Made -e/-E options of dump accept an unlimited number of inodes
194 to be excluded, rather than a hardcoded maximum. Thanks to
195 Dietrich Rothe <d-rothe@users.sourceforge.net> for the patch.
196
197 4. Updated the autoconf system to 2.50. Forced the -D_BSD_SOURCE
198 and -D_USE_BSD_SIGNAL defines in configure in order to solve
199 64bit build problems because quad_t is redifined with a
200 different signature. Thanks to Mike Harris <mharris@redhat.com>
201 for reporting this bug.
202
203 5. Made restore build on Solaris, making possible to
204 restore Linux's "enhanced" tapes. Thanks to Uwe Gohlke
205 <uwe@ugsoft.de> for the patch.
206
207 6. Made an extension in the dump tape format capable of saving
208 MacOSX specific inode extensions. Uwe Gohlke <uwe@ugsoft.de>
209 wrote the extension and contributed the restore code back
210 into this codebase. The same extension mechanism will be
211 used in the future to save ACLs...
212
213 7. Made rmt work correctly with regard to QFA and local/remote
214 files and tapes. The remote access will however work only
215 when the dump provided rmt version is used. If you want to
216 use another rmt server, please do not use the QFA feature.
217 Thanks to Uwe Gohlke <uwe@ugsoft.de> for the patch.
218
219 Changes between versions 0.4b33 and 0.4b34 (released April 18, 2003)
220 ====================================================================
221
222 1. Fixed the INSTALL file to reflect the actual install paths.
223 Thanks to David Ronis <ronis@ronispc.chem.mcgill.ca> for
224 reporting the bug.
225
226 2. Fixed the configure script to only check for headers presence
227 instead of trying to compile them. This should fix issues
228 with old build environments. Thanks to Kari Lempiainen
229 <kari@funky.pp.fi> for reporting the bug.
230
231 3. Fixed restore to correctly ignore sockets when comparing
232 a dump (as socket cannot be properly restored anyway). Thanks
233 to Gunther Reiszig <gunther@mit.edu> for reporting the bug.
234
235 4. Fixed restore to correctly access the archive file (-A argument)
236 even when using a remote tape. Thanks to Fabrice Bellet
237 <fabrice@bellet.info> for reporting the bug.
238
239 5. Fixed (again) handling of long (largefile) seeks in rmt.
240 Thanks to Fabrice Bellet <fabrice@bellet.info> for reporting
241 the bug.
242
243 6. Fixed restore corner case when dealing with large block sizes
244 dump is able to write now (-b 1024). Thanks to Fabrice Bellet
245 <fabrice@bellet.info> for reporting the bug.
246
247 7. Fixed a bug preventing dump to access a filesystem having
248 a label exactly 16 bytes in length. Thanks to <minduni@ti-edu.ch>
249 for reporting the bug.
250
251 8. Made dump store dump dates using explicit timezones, fixing a
252 problem with incremental dumps if the timezone is modified
253 between the dumps. Thanks to Philippe Troin <phil@fifi.org> for
254 the bug report and the patch.
255
256 9. Fixed a bug encountered when dumping individual files (not full
257 filesystems or directories) and dangling symbolic links happen
258 to be in the list of files. For as far as dump is concerned,
259 dangling symbolic links are allowed, and are dumped as is.
260 Thanks to Jin-su Ahn <jsahn@ee.snu.ac.kr> for reporting the
261 bug and providing the fix.
262
263 10. Fixed open and creation modes and permissions for QFA and
264 table-of-contents files in dump and restore. Thanks to
265 Philippe Troin <phil@fifi.org> for the patch.
266
267 11. Fixed the archive file descriptor handling enabling it to be 0.
268 This can happen in some cases when shell redirections are used.
269 Thanks to Philippe Troin <phil@fifi.org> for the patch.
270
271 12. Delayed the opening of archive file until after suid had been
272 dropped (fixing a possible security issue if dump is suid).
273 Thanks to Philippe Troin <phil@fifi.org> for the patch.
274
275 13. Fixed the 'S' command handling in the rmt client part.
276 Thanks to Philippe Troin <phil@fifi.org> for the patch.
277
278 14. Modified the end-of-tape script handling to print out statistics
279 (and stop the timer) before launching the eot script. Also, the eot
280 script does not get run anymore when using -M (which makes sense) or
281 when multiple tapes are listed on the command line
282 (-f tape0,tape1,tapen) (which also makes sense).
283 Thanks to Philippe Troin <phil@fifi.org> for the patch.
284
285 15. Relicensed dump/restore under the 'revised' BSD license, as per
286 ftp://ftp.cs.berkeley.edu/ucb/4bsd/README.Impt.License.Change.
287
288 16. Added LZO compression to dump. This new compression method has
289 the advantage of being super fast, thus not killing tape streaming
290 on slow machines. Thanks to Helmut Jarausch
291 <jarausch@igpm.rwth-aachen.de> for the patch and to
292 Markus Oberhumer <markus@oberhumer.com> for giving special permission
293 to include his miniLZO project (GPL licensed) in dump/restore.
294
295 17. Some small buffer overruns fixes in rmt. Thanks to Antonomasia
296 <ant@notatla.demon.co.uk> for reporting the bugs.
297
298 18. Added a special rmt version which can do encryption when writing
299 to tape. Read examples/encrypted_rmt/README for details on
300 how to enable and configure it. Thanks to Ken Lalonde
301 <ken@globalremit.com> for the patch.
302
303 19. Made dump work with 2.5 kernel end of tape early warning semantics.
304 Thanks to Kai Makisara <Kai.Makisara@kolumbus.fi> for the patch.
305
306 20. Fixed a bug which caused dump -w|-W not to work anymore, because
307 the fs_freq and fs_passno fields in /etc/mtab are always set
308 to 0 0. Thanks to Trent Piepho <xyzzy@speakeasy.org> for
309 reporting the bug.
310
311 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
312 =======================================================================
313
314 1. Added a note in the restore man page clarifying the question
315 'set the permissions on the current directory ?' asked by
316 restore at the end of treatment in -i and -x modes.
317
318 2. Fixed the endianess issues when reading compressed tapes.
319 Thanks to Dark Force <daq4th@users.sourceforge.net> for
320 reporting this bug and providing test cases.
321
322 3. Fixed the "ACL won't be dumped" warning message (which showed
323 an extra, unrelated error message). Thanks to Dragan Krnic
324 <dkrnic@lycos.com> for reporting this bug.
325
326 4. Made dump look first into /etc/mtab, then into /etc/fstab
327 when searching for filesystem paths. Also fixed some problems
328 caused by binding mounts or multiple block device mounts.
329 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
330 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
331 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
332 and #160305, Sourceforge bugs #618699 and #679832).
333
334 5. Made dump's -I option accept the value '0' meaning all the
335 read errors will be ignored. This can be useful when running
336 dump from unattended sessions (like cron jobs). Thanks to
337 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
338
339 6. Fixed the output of dump to indicate 'blocks' instead of
340 'tape blocks' in the various messages (blocks are always
341 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
342 made it clearly print the current blocksize at the start of
343 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
344 the suggestions.
345
346 7. Made rmt understand long (largefiles) seeks.
347
348 8. Fixed build with very old versions of libext2fs, where
349 EXT2_FT_* constants were undefined.
350
351 9. Made dump accept the dumpdates path on the command line
352 (-D file option) instead of using only the hardcoded one.
353 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
354 suggestion.
355
356 10. Enabled rmt, LFS, readline, QFA options by default in
357 ./configure. Updated the configure process (new versions
358 of config.guess, config.sub etc).
359
360 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
361 =======================================================================
362
363 1. Changed dump to use fcntl(F_SETLK) style locking instead
364 of flock() when locking the dumpdates file. With the old
365 locking scheme, a local user having read rights on the
366 dumpdates file could be able to do a Denial of Service attack
367 on dump. In order to lock the dumpdates file with the new
368 scheme, the user would need to have write access on the file.
369 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
370 reporting the bug (originally a bugtraq post).
371
372 2. Fixed interactive 'ls' which caused spurious errors warnings
373 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
374 <jorgen_ostling@users.sourceforge.net> for reporting this
375 bug.
376
377 3. Fixed dump's estimate when dealing with sparse inodes.
378
379 4. Modified dump to allow setting a blocksize bigger than 32kB
380 (raised the limit to 1024kB), because newer hardware needs
381 this for proper operation. Thanks to Dirk Traenapp
382 <dtraenapp@users.sourceforge.net> for reporting this.
383
384 5. Fixed a bug causing Dump to stop and report an error if an
385 inode number in the exclude file was followed by some amount
386 of whitespace. Thanks to Jeffrey Sofferin
387 <sofferin@users.sourceforge.net> for reporting this bug.
388
389 6. Fixed a bug which caused restore, in some particular cases,
390 to ask some 'scary' questions and leave a bunch of RSTTMP
391 directories behind when restoring incremental tapes. Thanks
392 to Philippe Troin <phil@fifi.org> for reporting this bug and
393 providing the test cases.
394
395 7. Changed the wording when inodes are excluded from dump:
396 replaced 'Added inode 7 to exclude list' with
397 'Excluding inode 7 (resize inode) from dump', as suggested
398 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
399
400 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
401 ===================================================================
402
403 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
404 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
405 <eros@lamel.bo.cnr.it> for reporting the bug.
406
407 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
408 ===================================================================
409
410 1. Made dump print out the ext2 logical block number in case of a read
411 error, which can be given as an argument to debugfs' ncheck command,
412 in order to find out the affected inode. Added note about this
413 usage in dump's man page.
414
415 2. Fixed a problem in restore when reading tapes written on big
416 endian machines with very old versions of dump. The patch was
417 contributed by George Helffrich <george@geology.bristol.ac.uk>.
418
419 3. Fixed the tape length calculation when using large tapes
420 and compression. Thanks to Georg Lippold
421 <g_lippold@sourceforge.net> for reporting the bug.
422
423 4. Added a new set of examples in dump_on_cd_2 directory, based
424 on dump_on_cd examples but somewhat enhanced, supporting DVD
425 media, and localized in english and german. Thanks to
426 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
427
428 5. Made dump save 32 bit UID/GID. Internally, this was achieved
429 by switching from the old BSD inode format to the new BSD
430 inode format, which means that the tape format was changed.
431 However, since all restore versions out there should
432 transparently support both inode formats, the change should
433 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
434 reporting the bug.
435
436 6. Fixed a lot of warnings in the code shown when compiling
437 with 'gcc -W'. Thanks to Matthias Andree
438 <matthias.andree@stud.uni-dortmund.de> for reporting this.
439
440 7. Fixed a small markup bug in the dump man page. Thanks to
441 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
442 patch.
443
444 8. Rewrote entirely the man pages using the tmac.an macro
445 package (Linux man page format) instead of the original BSD
446 format. They should be now cleaner and easier to modify.
447
448 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
449 ==================================================================
450
451 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
452 non Linux clients were misinterpreted. The description of the problem
453 (incompatible numbering in Linux mtio opcodes) is documented at
454 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
455 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
456 providing an excellent, cross-platform replacement for rmt in his
457 star package.
458
459 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
460 which caused restore to fail when the inode maps were not entirely
461 contained on the first volume (when using really small volumes or
462 when dumping a huge number of inodes).
463
464 3. Fixed a problem in dump, where files in subdirectories of directories
465 that have the nodump flag set (or being excluded with -e/-E) were
466 sometimes incorrectly dumped (instead of being excluded). The original
467 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
468 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
469 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
470 help testing the patch.
471
472 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
473 which makes one able to pipe the output of dump, by the net, to
474 a remote CD-burner server.
475
476 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
477 and use GNU's symbolic syntax over rmt instead of numerical values
478 to assure multiple platform compatibility.
479
480 6. Documented the -d option in restore.
481
482 7. Added a -v (verbose) mode to dump. For now it just prints the number
483 of the inode being dumped, but this could evolve in future versions
484 to include interesting debugging output.
485
486 8. Added a -o flag to restore, which automatically restores the current
487 directory permissions in -i or -x mode, without asking the operator
488 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
489 and Peter Samuel <peters@e-smith.com>.
490
491 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
492 ====================================================================
493
494 1. Fixed a bug in the QFA feature which could have caused
495 a bad tape positionning on restore (causing extra delays in
496 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
497
498 2. Added a small note in the dump man page specifying that
499 there should be no white space between the option letter and
500 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
501
502 3. Made restore work with older versions of the readline library,
503 by checking in configure for several extended symbols. Restore
504 can now be compiled with a readline as old as the 2.0 release
505 (though it may be a good idea to upgrade it to a more recent
506 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
507 reporting the build failures.
508
509 4. Fixed a performance problem with the QFA file creation in
510 dump, which made unnecessary seeks on the tape slowing down
511 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
512 reporting this issue.
513
514 5. Removed the inclusion of some kernel headers in the dump
515 source, which prevented the compile in some kernel/glibc
516 headers/architecture combination. Thanks to Bdale Garbee
517 <bdale@gag.com> for reporting the bug.
518
519 6. Added the appropriate error message when dump fails to
520 open the output file for writing. Thanks to Amith Varghese
521 <amithv@yahoo.com> for reporting this bug.
522
523 7. Made restore able to understand large Solaris ufsdump tapes
524 (containing inodes bigger than 4194304). Sun have introduced
525 an "extension" to the dump tape format when dealing with
526 those inodes, which was uncorrectly handled by Linux restore.
527 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
528 providing a test case.
529
530 8. Added the -m parameter to dump which optimises the output for
531 inodes having been changed but not modified since the last dump
532 ('changed' and 'modified' have the meaning defined in stat(2)).
533 For those inodes, dump will save only the metadata, instead of
534 saving the entire inode contents. Inodes which are either
535 directories or have been modified since the last dump are saved
536 in a regular way. Uses of this flag must be consistent, meaning
537 that either every dump in an incremental dump set have the flag,
538 or no one has it.
539
540 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
541 =======================================================================
542
543 1. Fixed behaviour of dump when exceeding resource limits
544 (SIGXFSZ treatment).
545
546 2. Added the -L flag to restore to allow the user to specify a
547 maximal allowed number of miscompares when using restore
548 with the -C option to check the backup.
549
550 3. Detailed the manual entry for the -N option of restore.
551
552 4. Added the -a flag to restore to make able doing unattended
553 restores in -i or -x mode (automatically walks through the
554 multiple dump volumes).
555
556 5. Extended the QFA mode to work with local files and/or
557 remote tapes and files. This way, restore can know in advance
558 the tape number and the offset for the inodes to extract and
559 can minimize the extraction time by seeking directly to the
560 good tape/offset.
561
562 6. Added the -A <archive> option to both dump and restore,
563 which makes dump to archive a dump table-of-contents in
564 the specified file to be used by restore to determine
565 whether a file is in the dump file that is being restored.
566 (the archive file syntax is also compatible with the
567 Solaris ufsdump generated one).
568
569 7. Small fix in restore making it able to read some (broken ?)
570 Solaris ufsdump tapes.
571
572 8. Fixed dump to correctly recognise the root filesystem when using
573 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
574 for reporting this bug.
575
576 9. Added the -P <file> option to restore to create a
577 Quick File Access file from an already made dump. Patch
578 contributed by Uwe Gohlke <uwe@ugsoft.de>.
579
580 10. Made restore compile and run on Solaris, making it a
581 possible replacement for the standard ufsrestore. Port was
582 contributed by Uwe Gohlke <uwe@ugsoft.de>.
583
584 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
585 =====================================================================
586
587 1. Added a set of backup scripts from Eugenio Diaz
588 <getnito@yahoo.com> in the examples section. It features
589 automatic (cron based) full and incremental dumping of
590 several filesystems on a separate filesystem.
591
592 2. Fixed a off-by-one miscalculation which disabled dumping
593 a one letter subdirectory. Thanks to Chris Siebenmann
594 <cks@utcc.utoronto.ca> for reporting the bug.
595
596 3. Fixed several restore bugs occuring when trying to
597 restore some missing files on the tape. Thanks to Chris
598 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
599
600 4. Fixed --with-ldopts configure argument passing, installing from
601 a separate object directory, makefile cleanups contributed
602 by <splite@purdue.edu>.
603
604 5. Fix a bug which could caused, in some conditions, the highest
605 number inode of a filesystem, to not be dumped. Many thanks
606 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
607 chase this bug.
608
609 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
610 =======================================================================
611
612 1. Added a mini howto from Patrick Walsh in the examples
613 directory.
614
615 2. Minor man pages syntax corrections. Thanks to
616 Chris Danis <screechco@home.com> for reporting the bugs.
617
618 3. Added a script from David B. Peterson <dave@toppledwagon.com>
619 to the examples section. It features dumping several
620 filesystems to a remote tape drive upon ssh.
621
622 4. Added a patch provided by Richard Jones <rich@annexia.org>
623 which allows BRADEMAX (number of read errors tolerated by
624 dump) to be adjusted using the -I option.
625
626 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
627 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
628 patch.
629
630 6. Add the -l option to restore to specify if, when doing a
631 remote restore, the file used to access the backup is a
632 regular file (the defaults being a tape drive). Restore needs
633 to know this information when reading a remote compressed
634 dump. Previously, this information was autodetected, but
635 the autodetection code fails (with ioctl: Inappropriate ioctl
636 for device) when using a non Linux remote box. Thanks to
637 many users and especially to Eros Albertazzi
638 <eros@lamel.bo.cnr.it> for reporting this.
639
640 7. Found a workaroung for the dump deadlock problem (3 childs
641 stuck in pause(), father in read()). The workaround seems
642 to work for me and several beta-testers. If it doesn't work
643 for you, please report back.
644
645 8. Updated the RPM spec file (BuildPrereq, URL etc).
646
647 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
648 ========================================================================
649
650 1. Fixed the permissions of a newly created QFA file by dump.
651
652 2. Cleaned up the source of dump (the external variables
653 definition was a complete mess, making possible to have
654 objects overlap).
655
656 3. Fixed restore to use the full tape volume path when doing
657 a compare (since it changes the working directory to the
658 filesystem being compared in the process).
659
660 4. Added the -q option to dump which makes dump abort
661 whenever it needs operator attention. It should be
662 easier to use dump in scripts with this option.
663
664 5. Detect the use of incompatible options to dump and
665 refuse them (like -a and -B options together).
666
667 6. Added bzip2 compression to dump/restore (use option -j level
668 to select it). Note that you will need a restore version
669 >= 0.4b24 in order to restore a bzip2 compressed dump.
670 The same warning as for the zlib compression applies:
671 the tape format of a bzip2 dump is not compatible with the
672 original BSD tape format.
673
674 7. Fixed a overflow problem in dump corrupting the dump when
675 very large files were encountered. Thanks to Vladimir Nadvornik
676 for the bug report.
677
678 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
679 kernel buffer/page cache before starting the dump, helping
680 a bit those who use dump on mounted filesystems. Thanks to
681 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
682 for this suggestion.
683
684 9. Updated the RPM spec file following the RedHat changes
685 (dynamically linked binaries now in /usr/sbin etc).
686
687 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
688 which enables restore to recognise multi volume compressed dumps
689 done on CD. Included his scripts for dump (which pipe the dump
690 output directly into cdrecord) and restore. There is now
691 possible to to backups to CD on the fly!
692
693 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
694 ===================================================================
695
696 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
697 Marcus Meissner (Caldera International Security Dept.).
698
699 2. Implement the Sun rmt extended protocol. Patch submitted
700 by Ian Gordon <iangordon@users.sourceforge.net>.
701
702 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
703 the linux kernel header. This ensures that dump always has the
704 latest version of this file. Patch submitted by
705 Andreas Dilger <adilger@turbolinux.com>.
706
707 4. Report any filesystem present in either /etc/fstab with a
708 positive passno or /etc/dumpdates in dump -w output.
709 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
710
711 5. Fixed the looping problem in dump introduced in the
712 previous version.
713
714 6. Changed the -B option of dump to limit the size of
715 _compressed_ output per volume if compression is on.
716 Patch contributed by Helmut Jarausch
717 <jarausch@igpm.rwth-aachen.de>. Note however that, since
718 it is impossible to predict the size of the compressed
719 data before doing the compression, the -B limit is a bit
720 conservative.
721
722 7. Fixed a bug in reading the operator typed file/tape path for
723 the next volume in restore.
724
725 8. Implemented a "-F script" option to restore which permits the
726 user to specify a script which will be launched at the
727 beginning of each tape, useful for automatic programming of
728 tape changers for example. See the restore man page for the
729 script parameters and return codes.
730
731 9. Small fix for the QFA routines provided by Uwe Gohlke
732 <uwe@ugsoft.de>, and some recommendations for QFA uses in
733 the man pages.
734
735 10. Fixed the multivolume restoring where making a mistake
736 to the 'Mount next tape' prompt caused several blocks to
737 be lost.
738
739 11. Enhanced the -e option of dump to take as a parameter a
740 comma separated list of inode numbers.
741
742 12. Added the -E option to dump which specify a file containing
743 inode numbers to exclude from the dump.
744
745 13. Fixed the compressed multi-volume dump + restore.
746
747 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
748 ==================================================================
749
750 1. Made dump report the number of blocks dumped per volume.
751 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
752
753 2. Fix a bug in dump where waiting too much at the 'change volumes'
754 question caused the volume to be rewritten. Thanks to
755 Graham Reed <greed@users.sourceforge.net> for reporting the
756 bug and providing a patch.
757
758 3. Added a compression option to dump, contributed by Jerry
759 Peters <gapeters@worldnet.att.net>.
760
761 WARNING: the compression works only on files/pipes and
762 drives supporting variable block size.
763
764 WARNING2: tapes created using the compression feature are
765 incompatible with the standard dump tape format, and a
766 version of restore >= 0.4b22 will be needed for extraction.
767
768 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
769 architectures. Thanks to Paul Slootman <paul@debian.org> for
770 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
771 it upstream.
772
773 5. Many cleanups (CPP defines, const char warnings, check of
774 ext2fs COMPAT flags, time_t cleanups, added libext2 version
775 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
776
777 6. Made --prefix option work in configure. All the install path
778 are now based on the configure parameters.
779
780 7. Added the Quick File Access mode in dump/restore, contributed
781 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
782 a file tape position for each inode, and this file is used by
783 restore (if called with parameter Q and the filename)
784 to directly position the tape at the file restore is currently
785 working on. This saves hours when restoring single files from
786 large backups, saves the tapes and the drive's head. Use
787 --enable-qfa option of configure to compile in the QFA support.
788
789 8. Added the possibility to dump several files and directories
790 in a single invocation of dump. Thanks to Uwe Gohlke
791 <uwe@ugsoft.de> for implementing this option.
792
793 9. Fixed the dumping and restoring of files with holes
794 (files ending with holes didn't get dumped or restored
795 correctly).
796
797 10. Fixed a socket descriptor leak in dump, which leaved opened
798 3 file descriptors per dump process (and there is one dump
799 process per tape).
800
801 11. Fixed dump large file system support, by explicit use of
802 open64/lseek64/etc functions (explicit use needed because
803 e2fsprogs libraries don't behave well when compiled with
804 FILE_OFFSET_BITS=64).
805
806 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
807 ======================================================================
808
809 1. Fixed some bugs in the dump -w|-W logic introduced by
810 the previous version. Thanks to Andreas Dilger
811 <adilger@turbolinux.com> for his help on this one.
812
813 2. Fixed again a compilation problem when using old e2fs
814 headers (filesystem label related). Thanks to many users
815 who reported this stupid error.
816
817 3. Fixed a build problem on old lib5 based systems dealing with
818 _PATH_MNTTAB being undefined. Thanks to John Adams
819 <johna@onevista.com> for reporting the bug.
820
821 4. Improved the error detection in restore when creating
822 the temporary files in TMPDIR. Now restore will corectly
823 report a 'no space left on device' error instead of
824 strange errors which could imply an archive corruption.
825 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
826 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
827
828 5. Added the throughput information in dump's progression
829 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
830 for the patch.
831
832 6. Use libext2fs's inode scanning routines, which are particularly
833 robust when dealing with errors (especially when having some
834 bad blocks on the filesystem). Many thanks to Colin
835 <colin@horizon.com> for reporting the bug, investigating
836 the issues, finding a workaround, writing the patch and
837 fully testing it... (of course, if this patch will break
838 anything, he is to blame for :-)).
839
840 7. Made dump and restore LFS aware. Dump can dump a filesystem
841 containing large files, generate a large file on output and
842 restore can restore them. This feature is not enabled by
843 default, use --enable-largefile option of configure to enable
844 it (you need to have a LFS aware glibc though). Thanks to
845 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
846 and to Theodore T'so <tytso@valinux.com> for his always
847 useful thoughts.
848
849 8. Made dump ask upon a tape write error if it should rewrite
850 the current volume (assume this is a bad tape) or if it should
851 assume an end-of-tape condition (useful for tape drives which
852 behaves badly at the end of the tape). Thanks to Andreas
853 Dilger <adilger@turbolinux.com> for the suggestion.
854
855 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
856 =======================================================================
857
858 1. Fixed a small compilation problem due to a change
859 in the definintion of the struct sigaction in
860 glibc 2.0 / libc5. Thanks to Gunther Schlegel
861 <schlegel@riege.de> for reporting the bug and to
862 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
863 a fix.
864
865 2. Modified the label and uuid analysis in order to be
866 self-contained (independant of kernel/libc headers). This
867 should fix the compile with older kernel/libc headers and
868 will preserve the functionality. Thanks to Bernhard Erdmann
869 <bernhard.erdmann@gmx.de> for reporting the bug.
870
871 3. The 'exclude inode' option, if applied to a directory
872 inode, excludes now correctly all the leaves of this
873 directory. Thanks to John R. Dennison
874 <gerdesas@users.sourceforge.net> for reporting the bug.
875
876 4. Fixed the '-e' option to disable the possibility
877 to exclude the root inode (which causes the dumps to
878 be unreadable by restore). Prevented array overflow
879 when multiple -e options are used.
880
881 5. Fixed dump to correctly interpret a filesystem argument
882 which is a mountpoint and it is not an absolute path
883 (as specified in the fstab). Thanks to Bernhard R. Erdmann
884 <be@berdmann.de> for reporting the bug.
885
886 6. Made dump able to backup files larger than 2 GB. Note that
887 dump still doesn't cope with files larger than 4 GB.
888
889 7. Restore the real uid and gid before invoking an external
890 RSH program (big hole when dump or restore is suid root!).
891
892 8. Get the values for the default tape device and the location
893 of fstab file from the system headers. Thanks to
894 Andreas Dilger <adilger@turbolinux.com> for the patch.
895
896 9. Made dump -w|-W report all recognized filesystems
897 present in either /etc/fstab or /etc/dumpdates, and present
898 the list in the same order as in fstab file. Thanks
899 to Andreas Dilger <adilger@turbolinux.com> for the patch.
900
901 10. Made dump's -a (automatic end of tape detection) the
902 default. Specifying one of -B, -c, -d or -s options will
903 override the EOM detection. Thanks to Andreas Dilger
904 <adilger@turbolinux.com> for the patch.
905
906 11. Save the ext2 filesystem label into the dump volume label.
907 Specifying a volume label on the command line (-L option)
908 will override this feature. Thanks to Andreas Dilger
909 <adilger@turbolinux.com> for the patch.
910
911 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
912 =====================================================================
913
914 1. Fixed the signal handling in dump (which I broke in 0.4b17)
915 which was causing several strange dump failures (dump
916 hanged or segmentation faults etc).
917
918 2. Specified the default blocksize in the dump man page.
919
920 3. Changed two info messages of restore to be written on stdout
921 instead of stderr in order to leave stderr only for errors
922 or warnings. Thanks to Stephen Atwell
923 <satwell@urbana.css.mot.com> for the suggestion.
924
925 4. Corrected an off by one calculation which prevented
926 dumping a directory having a 1 character name length.
927 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
928 for reporting the bug.
929
930 5. Reinforce dump in order to allow only level 0 and no
931 -u option when dumping a subdirectory, and document
932 this in the man page. Thanks to Bernhard Erdmann
933 <bernhard.erdmann@gmx.de> for reporting the bug.
934
935 6. Fixed a small harmless bug in restore which caused
936 in some conditions hard links to be restored several
937 times (therefore generation some warning messages).
938 Thanks to Kenneth Porter <shiva@well.com> for
939 reporting the bug.
940
941 7. Updated the RPM spec file to the latest RedHat version,
942 providing FHS packaging and other cosmetic changes.
943 You will need RPM version >= 3.0.5 in order to build the RPM.
944
945 8. Updated the configure script to check for libtermcap
946 before checking for libreadline, because we need this
947 library in order to compile the readline support.
948
949 9. Made dump understand the LABEL= and UUID= notation
950 both in /etc/fstab and on the command line. Note that
951 you will need the /proc filesystem in order to use
952 these notations. Thanks to Erik Troan <ewt@redhat.com>
953 for providing the patch.
954
955 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
956 ===================================================================
957
958 1. Fixed a potential buffer overflow in restore. Thanks
959 to Stan Bubrouski <satan@fastdial.net> for reporting
960 the bug.
961
962 2. Fixed a readline-related bug which prevented
963 'cat DUMPFILE | restore -i -f -' from working. Thanks
964 to Charles Karney <karney@users.sourceforge.net>
965 for the bug report.
966
967 3. Changed a few "panic" into "exit", causing restore to
968 be more stable against some attacks (like the last one
969 reported on Bugtraq, although the last version of restore
970 was not vulnerable - just dumped core). Thanks to
971 Andreas Hasenack <andreas@conectiva.com.br> for reporting
972 the bugs to me.
973
974 4. Removed the suid-root bit on dump and restore in the
975 default build (and generated RPMs). It should be safer
976 now. Users who need the suid-root capabilities in order
977 to make network backups should read first the man page
978 of dump and enable the suid bit manually.
979
980 5. Added -ltermcap to the compile parameters for restore
981 when using readline support, in order to make the compile
982 process work on some systems (SuSE ?). Thanks to
983 Patrik Schindler <poc@pocnet.net> for reporting the bug.
984
985 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
986 ====================================================================
987
988 1. The -F script is called now *only* at the end of a tape,
989 not at the end of the dump. Thanks to Christian Haul
990 <haul@informatik.tu-darmstadt.de> for the bug report.
991
992 Normally, the device name and the current volume number
993 are passed on the command line of the script. However,
994 if you want the old style script (with no arguments
995 passed on the command line), you can enable it in
996 configure with the --enable-oldstylefscript.
997
998 2. Use posix signal handling to preserve dump functionality
999 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
1000 the patch.
1001
1002 3. Made the exit code of restore in case of a 'C'ompare
1003 command reflect the result of the compare. An exit status
1004 of 0 means the dump archive is correct, 1 means tape errors,
1005 2 means that some files were modified. Thanks to Kenneth Porter
1006 <shiva@well.com> for the suggestion.
1007
1008 4. Made (finally) quotes work in interactive restore.
1009
1010 5. Small fixes in order to allow dump to compile with a
1011 really old e2fsprogs version. Thanks to Ian Zimmerman
1012 <itz@speakeasy.org> for the bug report.
1013
1014 6. Add GNU readline capabilities to interactive restore.
1015 Use configure's --enable-readline flag to enable this feature.
1016 Thanks to Patrik Schindler <poc@pocnet.net> for the
1017 suggestion.
1018
1019 7. Do the compare on the fly in restore 'C' mode (this will
1020 allow not to exhaust the available /tmp space when
1021 ccmparing large files). Thanks to Kenneth Porter
1022 <shiva@well.com> for the suggestion.
1023
1024 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
1025 ====================================================================
1026
1027 1. Fixed some several duplicate 'DUMP: DUMP:' in the
1028 output of dump.
1029
1030 2. Corrected the estimation of blocks to dump. Note that
1031 this is still wrong for large directory inodes, due
1032 to the size of a BSD directory entry on the tape
1033 (estimation is based on the size of the ext2 directory,
1034 which is smaller because it doesn't contain the null
1035 character at the end).
1036
1037 3. Made dump report the total number of MB written to
1038 tape at the end of run. Thanks to W. Reilly Cooley
1039 <wcooley@nakedape.cc> for the patch.
1040
1041 4. Added the -X option to restore giving the possibility
1042 to read the names of the files to be extracted/listed
1043 from a text file (in addition of the command line).
1044 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
1045 patch.
1046
1047 5. Added the device name and the current volume number
1048 as arguments to the end of tape script (dump -F option).
1049
1050 6. Made the multi-volume dump work again (after having
1051 broken it in 0.4b15).
1052
1053 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
1054 ===================================================================
1055
1056 1. Added a prompt command in interactive restore mode. Thanks
1057 to Andreas Dilger <adilger@home.com> for the patch.
1058
1059 2. Fixed a buffer overflow problem in dump (caused by
1060 not checking the size of the filesystem parameter).
1061 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
1062 reporting this on Bugtraq (and to several dump users
1063 who forwarded me his mail).
1064
1065 3. Added the '-F script' option to dump in order to
1066 launch a script at the end of each tape (to be used
1067 with a tape changer, or to notify the sysadmin by
1068 pager etc.).
1069
1070 4. Fixed a bug in restore compare code caused by the changes
1071 I made in 0.4b14.
1072
1073 5. Fixed the treatment of options using the old BSD syntax
1074 in both dump and restore.
1075
1076 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
1077 =======================================================================
1078
1079 1. Fixed a bug in dump which may have caused invalid deleted
1080 directories to be dumped out if they were deleted after the
1081 mapping pass. This could occure on active filesystem and lead
1082 to heap corruption (causing dump malfunction in many possible ways).
1083 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
1084 bug and submitting the patch.
1085
1086 2. Fixed the handling of the filesystem parameter in dump. A
1087 '/mnt/disk' parameter caused the disk contents to be dumped,
1088 but a '/mnt/disk/' parameter caused the mountpoint directory
1089 to be dumped (generally an empty directory).
1090
1091 3. Improved the output of dump in order to tell which directory
1092 it is currently dumping (when dumping a subtree).
1093
1094 4. Added the '-e' exclude inode option to dump. Thanks to
1095 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
1096
1097 5. Added a REPORTING-BUGS file in order to provide a guide
1098 on how to correctly report a bug in dump/restore.
1099
1100 6. Fixed a restore problem when restoring a hard link to an inode
1101 having the immutable or append-only attribute set. Thanks to
1102 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
1103
1104 7. Fixed a compatibility problem between dumps made on little
1105 endian machines (the format was correct) and big endian
1106 machines (the format was incorrect). This fix break the
1107 compatibility with the older dumps made on big endian
1108 machines (sparc, m86k, ppc etc). For the first time in
1109 linux dump's history (I believe), the dumps made by *BSD,
1110 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
1111 so interchangeable. Thanks to Rob Cermak
1112 <cermak@ahab.rutgers.edu> for submitting the bug and
1113 helping me test the fix.
1114
1115 8. Fixed the way dump reports the remaining percent/time, if
1116 the number of blocks actually dumped exceeds the estimated
1117 number of blocks. Thanks to Jean-Paul van der Jagt
1118 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
1119
1120 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
1121 ======================================================================
1122
1123 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
1124 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
1125
1126 2. Removed the requirement to build the RPM as root from the
1127 spec file. Thanks to Christian Weisgerber
1128 <naddy@mips.rhein-neckar.de> for submitting this.
1129
1130 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
1131 causing dump to block when dumping really huge filesystems.
1132 Many thanks to Patrik Schindler <poc@pocnet.net> for
1133 helping me find this bug.
1134
1135 4. Fixed the treatment for an interrupt signal when dump access
1136 the remote tape through RSH. Thanks to Christian Weisgerber
1137 <naddy@mips.rhein-neckar.de> for providing the patch.
1138
1139 5. Fixed a bug which was causing dump/restore to display
1140 garbage characters instead of the remote host name.
1141
1142 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
1143 =====================================================================
1144
1145 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
1146 <kukuk@suse.de> for submitting the patch.
1147
1148 2. Fix for the exit code when using the size estimate option of
1149 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
1150 submitting the patch.
1151
1152 3. Handle EINTR in atomical reads/writes in dump, which was causing
1153 dump to fail on some systems. Thanks to Eric Jergensen
1154 <eric@dvns.com> for reporting the bug and submitting the patch.
1155
1156 4. Handle more than 16 characters for the device names in dumpdates.
1157 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
1158 tracking down the problem and proposing the solution.
1159
1160 5. Fixed a bug in dump which prevented the creation of the
1161 dumpdates file when doing a 0-level dump without already
1162 having a dumpdates file. Thanks to Patrik Schindler
1163 <poc@pocnet.net> for reporting the bug.
1164
1165 6. Changed the way dump 'S' flag reports the size estimate
1166 from number of blocks into bytes (making it compatible
1167 with the Solaris version, and simplifying things for
1168 amanda users). Thanks to Jason L Tibbitts III
1169 <tibbs@math.uh.edu> for reporting the bug.
1170
1171 7. Fixed a compatibility problem in linux/alpha dump tape format.
1172 Now the linux/alpha dump are (again) compatible with the
1173 other dump formats. But this breaks compatibility with
1174 older dumps made on alpha. Thanks to Mike Tibor
1175 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
1176
1177 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
1178 ======================================================================
1179
1180 1. Added a '--enable-kerberos' to configure.
1181
1182 2. Added a 'S' option to dump which determines the amount of space
1183 that is needed to perform the dump without actually doing it, similar
1184 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
1185 <cermak@ahab.rutgers.edu>.
1186
1187 3. Added a 'M' multi-volume option to dump and restore which enables
1188 dumping to multiple files (useful when dumping to an ext2
1189 partition to make several dump files in order to bypass the 2GB
1190 file size limitation). The 'f' argument is treated as a prefix and
1191 the output files will be named <prefix>001, <prefix>002 etc. With
1192 the 'M' flag, restore automatically selects the right file without
1193 asking to enter a new tape each time.
1194
1195 4. Fixed a memory leak which was causing dump to grow very big
1196 (270MB when dumping a 10GB filesystem...). Thanks to Jason
1197 Fearon <jasonf@netrider.org.au> for reporting the bug.
1198
1199 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
1200 ======================================================================
1201
1202 1. Make configure test if the system glob routines support
1203 extended syntax (ALTDIRFUNC). If not, use the internal glob
1204 routines instead of system ones. Thanks to Bernhard Sadlowski
1205 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
1206 and helping me resolve this and other minor libc5 compiling
1207 glitches.
1208
1209 2. Fix a problem when dumping a ext2fs with the 'filetype'
1210 feature enabled. Thanks to Patrick J. LoPresti
1211 <patl@cag.lcs.mit.edu> for reporting the bug and to
1212 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
1213
1214 3. Made the nodump flag work on directories. A directory which
1215 has the nodump flag gets never dumped, regardless of its
1216 contents.
1217
1218 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
1219 which allows dump on an active ext3 filesystem. However, this
1220 is a "quick and dirty" patch which enables backup of an ext3
1221 filesystem through the ext2 compatibility (by ignoring the
1222 NEEDS_RECOVERY bit). The journal file is not recognized and
1223 it is dumped (it should not).
1224
1225 5. Test the superblock compatibility flags when dumping, in order
1226 to be sure that we know how to deal with specific features.
1227
1228 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
1229 ====================================================================
1230
1231 1. Use lchown instead of chown, fixing a possible security problem
1232 when restoring symlinks (a malicious user could use this
1233 to deliberately corrupt the ownership of important system files).
1234 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
1235 this and providing the patch.
1236
1237 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
1238 ====================================================================
1239
1240 1. Put dump sources under CVS, added Id tags in all files so
1241 one can use 'ident' on binary files.
1242
1243 2. Added the dump/restore version in the usage text so one can
1244 easily verify the version he is using.
1245
1246 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
1247 a va_start/va_end problem on linux-ppc (always call va_start
1248 va_end in pairs each time we use a vararg function).
1249
1250 4. Added again the DT_* constants because old libc does not
1251 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
1252 for submitting the bug report.
1253
1254 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
1255 this should enable dumping big (huge) filesystems.
1256
1257 6. Added the RSH environment variable in order to be able to
1258 use a rsh replacement like ssh when doing remote backups (and
1259 bypass the security limitations of rcmd). Now you can do remote
1260 backups without being root (or making dump setuid root).
1261
1262 7. Modified again the way dumpdates works. For incremental dumps,
1263 we need to read dumpdates even if we are not using 'u' option.
1264 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
1265 this should work.
1266
1267 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
1268 ===================================================================
1269
1270 1. Removed the 'k' flag from the restore 'about' text if kerberos
1271 was not compiled in.
1272
1273 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
1274 to them (fsetflags takes a char*, setflags an open fd!).
1275
1276 3. (f)setflags is called only if the flags aren't empty. If the
1277 file is a special file, a warning is printed, because changing
1278 flags implies opening the device. Normally, a special file
1279 should not have any flag... (Debian bug #29775, patch provided
1280 by Abhijit Dasgupta <abhijit@ans.net>).
1281
1282 4. Made possible to dump a file system not mentioned in /etc/fstab.
1283 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
1284
1285 5. Changed the default behaviour to not create dumpdates
1286 unless 'u' option is specified. Removed the old "debian-patch"
1287 which provided the same thing. (Debian bug #38136, #33818).
1288
1289 6. Removed all those dump*announce, since they were getting old...
1290
1291 7. Added warning messages if dumpdates does not exist and
1292 when an inferior level dump does not exist (except for a level 0
1293 dump).
1294
1295 8. Debugged the glob calls in interactive mode: restore used a
1296 dirent struct which was different from the /usr/include/dirent.h
1297 one (this used to work, is it a glibc2 change?), so none of the
1298 compat glob (which used /usr/include/dirent.h) or the system glob
1299 worked. Restore use now the system dirent (and the system
1300 DT_* constants), which are compatible with BSD ones.
1301
1302 9. Added a configure flag (--with-dumpdatespath) to specify
1303 the location of dumpdates file. By default, it is
1304 /etc/dumpdates.
1305
1306 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1307 included the current date/version in man pages.
1308
1309 11. Corrected the estimation of remaining time when
1310 the operator doesn't change the tapes quickly enough. This
1311 was an old bug, I thought I corrected it, and discovered
1312 that in fact it was corrected in two different places, so
1313 the results canceled each other...
1314
1315 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1316 ===================================================================
1317
1318 1. Integrated multiple patches from RedHat, Debian and SuSE:
1319
1320 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1321 isn't specified.
1322 - several fixes in the man pages.
1323 - update the default tape device to /dev/st0.
1324 - many updates for Linux Alpha (byte ordering, size_t etc).
1325 - buffer overruns.
1326 - use environment variable for TMPDIR (instead of /tmp).
1327 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1328 - workaround egcs bug (RedHat bugs #4281 and #2989).
1329 - wire $(OPT) throughout Makefile's.
1330
1331 2. Upgrade the dump revision to 1, making possible to dump filesystems
1332 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1333
1334 3. Fix some compile warnings, prototype all functions.
1335
1336 4. Use glibc err/glob instead of internal compatibility
1337 routines (only if available).
1338
1339 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1340 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1341
1342 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1343 ======================================================================
1344
1345 1. Integrated the changes from FreeBSD-3.1-RELEASE
1346 (mostly bug fixes, buffer overruns, dump has now an "automatic
1347 tape length calculation" flag, dump/restore can use kerberos now
1348 (this is NOT tested), use environment variables for TAPE and
1349 RMT etc.).
1350
1351 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1352
1353 3. Corrected a bug in restore when using 'C' option with multi-volumes
1354 tapes (files splited accros two tapes give "size changed" errors
1355 when comparing).
1356
1357 4. Corrected the long standing bug when dumping multiple tapes.
1358 This works for me, needs further testing.
1359
1360 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1361 ====================================================================
1362
1363 1. Dump now runs correctly on kernels 2.1.x
1364 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1365
1366 Changes between versions 0.4b2 and 0.4b3
1367 ========================================
1368
1369 1. Use realpath() if available
1370
1371 2. Report statistics
1372
1373 Changes between versions 0.4b1 and 0.4b2
1374 ========================================
1375
1376 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1377 the patch)
1378
1379 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1380
1381 3. Fixed configure.in to correctly handle --enable-rmt
1382
1383 Changes between versions 0.3 and 0.4b1
1384 ======================================
1385
1386 1. Integrated the changes from 4.4BSD-Lite2
1387
1388 2. Integrated the patches from Debian and Red Hat
1389
1390 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1391
1392 4. Changed dump to use the Ext2fs library to get block addresses. This
1393 should solve the endianness problem on SparcLinux.
1394
1395 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1396 distribution's one) to use autoconf
1397
1398 6. Fixed a few minor bugs
1399
1400 Changes between versions 0.2e and 0.2f
1401 ======================================
1402
1403 1. Added the creation of named pipes (fifos) in restore.
1404
1405 2. Added the -N flag in the restore manual page.
1406
1407 3. Added the file kernel.patch which contains the llseek() optimization
1408 patch for 1.2.x kernels.
1409
1410 4. Fixed a bug in the restoration of symbolic links: owner and group were
1411 not restored.
1412
1413 5. Integrated some changes from FreeBSD 2.2.
1414
1415 6. Added a call to ftruncate() after restoring each file to restore
1416 correctly files ending by a hole.
1417
1418 Changes between versions 0.2d and 0.2e
1419 ======================================
1420
1421 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1422 with owner = root (0) and group = root/wheel/whatever (0).
1423
1424 Changes between versions 0.2c and 0.2d
1425 ======================================
1426
1427 1. Dump is now able to backup 2GB+ filesystems.
1428
1429 2. Dump and restore can now be linked as static binaries.
1430
1431 Changes between versions 0.2b and 0.2c
1432 ======================================
1433
1434 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1435
1436 Changes between versions 0.2a and 0.2b
1437 ======================================
1438
1439 1. Really fixed the bug that I should have corrected in 0.2a.
1440
1441 2. Enabled optimization again.
1442
1443 Changes between versions 0.2 and 0.2a
1444 =====================================
1445
1446 1. Disabled the optimization during compilation.
1447
1448 Changes between versions 0.1 and 0.2
1449 ====================================
1450
1451 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1452 the fs_type field (actually, I modified the file fstab.c to make it
1453 use the mntent functions).
1454
1455 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1456 made by dump should be readable by the BSD restore and backups made
1457 by the BSD dump should be readable by restore. Unfortunately, this
1458 means that the dump format has changed between version 0.1 and version
1459 0.2 :-(
1460
1461 3. Dump is now able to backup a subtree, it is no longer limited to whole
1462 filesystems like the BSD version.
1463
1464 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1465 than 2 GB.
1466
1467 Changes between versions 0.0 and 0.1
1468 ====================================
1469
1470 1. Now create links rdump and rrestore during the `make install' step.
1471
1472 2. Linux port specific bugs added to the manual pages
1473
1474 3. Incorrect estimation of the number of tapes blocks fixed when doing
1475 incremental backups.
1476
1477 4. Better ls-like format in restore in interactive mode.