f56eb80cda6b53228074577cf2c7092dc5792102
[dump.git] / CHANGES
1 $Id: CHANGES,v 1.266 2005/01/14 13:01:32 stelian Exp $
2
3 Changes between versions 0.4b38 and 0.4b39 (released ???????????????)
4 =====================================================================
5
6 1. The newly added dump_on_cd_3 example was buggy, replace it
7 with an updated version from Andrew Basterfield
8 <bob@cemetery.homeunix.org>.
9
10 2. Made restore to chdir() back into the initial directory when
11 dumping core while aborting a comparision operation. The
12 previous behaviour was to write the corefile at the root of
13 the directory being compared, which could very well be
14 read only and preventing the corefile generation. Thanks
15 to Kenneth Porter <shiva@sewingwitch.com> for the bug report.
16
17 3. Silenced the failure to call fgetflags() when comparing an
18 entry which has no ext2 attributes (as in lsattr()).
19
20 Changes between versions 0.4b37 and 0.4b38 (released January 7, 2005)
21 =====================================================================
22
23 1. Fix a couple of troff syntax bugs in the man pages.
24 Thanks to Eric Raymond <esr@thyrsus.com> for the patch.
25
26 2. Made restore use either libncurses or libtermcap, depending
27 on which one is available at configure time.
28
29 3. Fixed restore negative size display bug when comparing a
30 dump containing files over 2GB. Thanks to Steve Bonds
31 <sbonds@users.sourceforge.net> for the bug report.
32
33 4. Do not save directory entries to non-dumped inodes
34 (excluded from dump). This will eliminate the 'missing
35 file' warnings when doing 'restore -C'.
36
37 5. Fix dump crash when backuping a huge (2TB) filesystem,
38 due to a bogus calculation on the inode map size.
39 Thanks to Kevin B. Haines <K.B.Haines@rl.ac.uk> for
40 submitting the bug and testing the fix.
41
42 6. Fix a problem in restore where the final \0 in the symbolic
43 link names could have been lost, generating corrupt filenames.
44 Thanks to Kyle Wilson <kyle.wilson@amd.com> for reporting the
45 bug.
46
47 7. Implemented a hash list for the directory names in restore.
48 The linear list used before caused problems in interactive
49 restores when dealing with directories having thousands of
50 entries. Thanks to Brian Ristuccia <bristuccia@starentnetworks.com>
51 for reporting the bug.
52
53 8. Improved restore -C, this time including the directory
54 attributes into the comparision.
55
56 9. Made restore understand tapes containing EA/ACLs (which will
57 be dumped by the next version of dump). In this version
58 extended attributes on the tape are ignored, for full EA/ACL
59 support wait for the next version or try the experimental EA
60 patch.
61
62 Changes between versions 0.4b36 and 0.4b37 (released July 7, 2004)
63 ==================================================================
64
65 1. Added the --enable-staticz configure option which enables
66 dump and restore to be linked with static versions of
67 libz and libbz2 (and dynamic versions of all other libraries).
68 This will make Debian users happy, because libz and libbz2
69 were the only needed libraries living in /usr, all the
70 others live in /lib. In case of system emergency, it is
71 better not to have to rely on an extra filesystem.
72 Thanks to Bdale Garbee <bdale@gag.com> for the suggestion.
73
74 2. Fix compilation on (at least the Linux Debian port to) AMD64.
75 (<ext2fs/ext2_types.h> defines some types (__s64 and __u64)
76 that are also defined by <linux/types.h> (<asm/types.h>) and
77 they conflict).
78
79 3. Make dump's reading of the dumpdates file a bit more robust,
80 preventing dump from crashing when the dumpdates file has
81 been modified by hand.
82
83 4. Fixed some offset calculations in dump code which could
84 lead to "bread lseek errors" on large filesystems. Thanks
85 to Bruce Lowekamp <lowekamp@users.sourceforge.net> for
86 reporting this bug and debugging the issue.
87
88 5. Made dump use the blkid library when searching for devices
89 by label or uuid instead of dump's own routines.
90
91 6. Corrected a bug in dump where a wrong LABEL=... line in
92 /etc/fstab could prevent dump from dumping unrelated
93 filesystems. Thanks to Bruce Lowekamp
94 <lowekamp@users.sourceforge.net> for reporting the bug.
95
96 Changes between versions 0.4b35 and 0.4b36 (released April 21, 2004)
97 ====================================================================
98
99 1. Fixed dump compilation with old gcc versions. Thanks to
100 Mike Castle <dalgoda@users.sourceforge.net> for the patch.
101
102 2. Fixed some warnings (howmany, roundup, powerof2 redefined)
103 when compiling against a recent glibc version.
104
105 3. Fixed a bug in restore preventing the read of a dump tape
106 written with Solaris 7 ufsdump. Thanks to Patrick Higgins
107 <phiggins@transzap.com> for reporting the bug and providing
108 the test case.
109
110 4. Changed dump to enable the creation of volumes of different
111 sizes in a single run (make -B accept a list of values).
112 Patch contributed by Florian Zumbiehl <florz@gmx.de>.
113
114 5. Use the glibc provided minor() and major() macros instead
115 of our own bitmask implementation. This should be safe for
116 when the major/minor namespace will migrate to 32 bits.
117 Thanks to Zhang Jun <zhangjun@nanjing-fnst.com> for reporting
118 the bug.
119
120 6. Made explicit in the dump man page that dump will not create
121 a remote file, it will only write to an already existing one.
122
123 7. Another try at making size estimates better again.
124
125 8. Put back the inconditional running of the end-of-tape
126 script which was changed in 0.434 to be run only when -M
127 or multiple -f were NOT specified. Some users rely on this
128 feature even when it is combined with -M/-f.
129
130 9. Fixed restore when restoring huge backups (where rstdir...
131 temporary files are over 2GB). Thanks to
132 Raphael Jaffey <rjaffey@artic.edu> for reporting this,
133 debugging the issue and making the patch.
134
135 10. Made restore understand FreeBSD UFS2 tapes. Thanks to
136 David <vrtin@users.sourceforge.net> for submitting the bug
137 and providing a test case.
138
139 11. Made dump work with an arbitrary integer level (no more 10
140 levels only limitation). Thanks to Matthew
141 <msvincen@midway.uchicago.edu> for the patch.
142
143 Changes between versions 0.4b34 and 0.4b35 (released December 21, 2003)
144 =======================================================================
145
146 1. Added a note in the dump man page saying that the default
147 blocksize can be 32 if -d is used with a high density argument.
148 Thanks to Antonios Christofides <A.Christofides@itia.ntua.gr>
149 for the patch.
150
151 2. Fixed configure to correctly understand CPPFLAGS, CFLAGS,
152 LDFLAGS environment variables. Thanks to Arcady Genkin
153 <antipode@users.sourceforge.net> for reporting the bug.
154
155 3. Made -e/-E options of dump accept an unlimited number of inodes
156 to be excluded, rather than a hardcoded maximum. Thanks to
157 Dietrich Rothe <d-rothe@users.sourceforge.net> for the patch.
158
159 4. Updated the autoconf system to 2.50. Forced the -D_BSD_SOURCE
160 and -D_USE_BSD_SIGNAL defines in configure in order to solve
161 64bit build problems because quad_t is redifined with a
162 different signature. Thanks to Mike Harris <mharris@redhat.com>
163 for reporting this bug.
164
165 5. Made restore build on Solaris, making possible to
166 restore Linux's "enhanced" tapes. Thanks to Uwe Gohlke
167 <uwe@ugsoft.de> for the patch.
168
169 6. Made an extension in the dump tape format capable of saving
170 MacOSX specific inode extensions. Uwe Gohlke <uwe@ugsoft.de>
171 wrote the extension and contributed the restore code back
172 into this codebase. The same extension mechanism will be
173 used in the future to save ACLs...
174
175 7. Made rmt work correctly with regard to QFA and local/remote
176 files and tapes. The remote access will however work only
177 when the dump provided rmt version is used. If you want to
178 use another rmt server, please do not use the QFA feature.
179 Thanks to Uwe Gohlke <uwe@ugsoft.de> for the patch.
180
181 Changes between versions 0.4b33 and 0.4b34 (released April 18, 2003)
182 ====================================================================
183
184 1. Fixed the INSTALL file to reflect the actual install paths.
185 Thanks to David Ronis <ronis@ronispc.chem.mcgill.ca> for
186 reporting the bug.
187
188 2. Fixed the configure script to only check for headers presence
189 instead of trying to compile them. This should fix issues
190 with old build environments. Thanks to Kari Lempiainen
191 <kari@funky.pp.fi> for reporting the bug.
192
193 3. Fixed restore to correctly ignore sockets when comparing
194 a dump (as socket cannot be properly restored anyway). Thanks
195 to Gunther Reiszig <gunther@mit.edu> for reporting the bug.
196
197 4. Fixed restore to correctly access the archive file (-A argument)
198 even when using a remote tape. Thanks to Fabrice Bellet
199 <fabrice@bellet.info> for reporting the bug.
200
201 5. Fixed (again) handling of long (largefile) seeks in rmt.
202 Thanks to Fabrice Bellet <fabrice@bellet.info> for reporting
203 the bug.
204
205 6. Fixed restore corner case when dealing with large block sizes
206 dump is able to write now (-b 1024). Thanks to Fabrice Bellet
207 <fabrice@bellet.info> for reporting the bug.
208
209 7. Fixed a bug preventing dump to access a filesystem having
210 a label exactly 16 bytes in length. Thanks to <minduni@ti-edu.ch>
211 for reporting the bug.
212
213 8. Made dump store dump dates using explicit timezones, fixing a
214 problem with incremental dumps if the timezone is modified
215 between the dumps. Thanks to Philippe Troin <phil@fifi.org> for
216 the bug report and the patch.
217
218 9. Fixed a bug encountered when dumping individual files (not full
219 filesystems or directories) and dangling symbolic links happen
220 to be in the list of files. For as far as dump is concerned,
221 dangling symbolic links are allowed, and are dumped as is.
222 Thanks to Jin-su Ahn <jsahn@ee.snu.ac.kr> for reporting the
223 bug and providing the fix.
224
225 10. Fixed open and creation modes and permissions for QFA and
226 table-of-contents files in dump and restore. Thanks to
227 Philippe Troin <phil@fifi.org> for the patch.
228
229 11. Fixed the archive file descriptor handling enabling it to be 0.
230 This can happen in some cases when shell redirections are used.
231 Thanks to Philippe Troin <phil@fifi.org> for the patch.
232
233 12. Delayed the opening of archive file until after suid had been
234 dropped (fixing a possible security issue if dump is suid).
235 Thanks to Philippe Troin <phil@fifi.org> for the patch.
236
237 13. Fixed the 'S' command handling in the rmt client part.
238 Thanks to Philippe Troin <phil@fifi.org> for the patch.
239
240 14. Modified the end-of-tape script handling to print out statistics
241 (and stop the timer) before launching the eot script. Also, the eot
242 script does not get run anymore when using -M (which makes sense) or
243 when multiple tapes are listed on the command line
244 (-f tape0,tape1,tapen) (which also makes sense).
245 Thanks to Philippe Troin <phil@fifi.org> for the patch.
246
247 15. Relicensed dump/restore under the 'revised' BSD license, as per
248 ftp://ftp.cs.berkeley.edu/ucb/4bsd/README.Impt.License.Change.
249
250 16. Added LZO compression to dump. This new compression method has
251 the advantage of being super fast, thus not killing tape streaming
252 on slow machines. Thanks to Helmut Jarausch
253 <jarausch@igpm.rwth-aachen.de> for the patch and to
254 Markus Oberhumer <markus@oberhumer.com> for giving special permission
255 to include his miniLZO project (GPL licensed) in dump/restore.
256
257 17. Some small buffer overruns fixes in rmt. Thanks to Antonomasia
258 <ant@notatla.demon.co.uk> for reporting the bugs.
259
260 18. Added a special rmt version which can do encryption when writing
261 to tape. Read examples/encrypted_rmt/README for details on
262 how to enable and configure it. Thanks to Ken Lalonde
263 <ken@globalremit.com> for the patch.
264
265 19. Made dump work with 2.5 kernel end of tape early warning semantics.
266 Thanks to Kai Makisara <Kai.Makisara@kolumbus.fi> for the patch.
267
268 20. Fixed a bug which caused dump -w|-W not to work anymore, because
269 the fs_freq and fs_passno fields in /etc/mtab are always set
270 to 0 0. Thanks to Trent Piepho <xyzzy@speakeasy.org> for
271 reporting the bug.
272
273 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
274 =======================================================================
275
276 1. Added a note in the restore man page clarifying the question
277 'set the permissions on the current directory ?' asked by
278 restore at the end of treatment in -i and -x modes.
279
280 2. Fixed the endianess issues when reading compressed tapes.
281 Thanks to Dark Force <daq4th@users.sourceforge.net> for
282 reporting this bug and providing test cases.
283
284 3. Fixed the "ACL won't be dumped" warning message (which showed
285 an extra, unrelated error message). Thanks to Dragan Krnic
286 <dkrnic@lycos.com> for reporting this bug.
287
288 4. Made dump look first into /etc/mtab, then into /etc/fstab
289 when searching for filesystem paths. Also fixed some problems
290 caused by binding mounts or multiple block device mounts.
291 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
292 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
293 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
294 and #160305, Sourceforge bugs #618699 and #679832).
295
296 5. Made dump's -I option accept the value '0' meaning all the
297 read errors will be ignored. This can be useful when running
298 dump from unattended sessions (like cron jobs). Thanks to
299 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
300
301 6. Fixed the output of dump to indicate 'blocks' instead of
302 'tape blocks' in the various messages (blocks are always
303 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
304 made it clearly print the current blocksize at the start of
305 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
306 the suggestions.
307
308 7. Made rmt understand long (largefiles) seeks.
309
310 8. Fixed build with very old versions of libext2fs, where
311 EXT2_FT_* constants were undefined.
312
313 9. Made dump accept the dumpdates path on the command line
314 (-D file option) instead of using only the hardcoded one.
315 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
316 suggestion.
317
318 10. Enabled rmt, LFS, readline, QFA options by default in
319 ./configure. Updated the configure process (new versions
320 of config.guess, config.sub etc).
321
322 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
323 =======================================================================
324
325 1. Changed dump to use fcntl(F_SETLK) style locking instead
326 of flock() when locking the dumpdates file. With the old
327 locking scheme, a local user having read rights on the
328 dumpdates file could be able to do a Denial of Service attack
329 on dump. In order to lock the dumpdates file with the new
330 scheme, the user would need to have write access on the file.
331 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
332 reporting the bug (originally a bugtraq post).
333
334 2. Fixed interactive 'ls' which caused spurious errors warnings
335 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
336 <jorgen_ostling@users.sourceforge.net> for reporting this
337 bug.
338
339 3. Fixed dump's estimate when dealing with sparse inodes.
340
341 4. Modified dump to allow setting a blocksize bigger than 32kB
342 (raised the limit to 1024kB), because newer hardware needs
343 this for proper operation. Thanks to Dirk Traenapp
344 <dtraenapp@users.sourceforge.net> for reporting this.
345
346 5. Fixed a bug causing Dump to stop and report an error if an
347 inode number in the exclude file was followed by some amount
348 of whitespace. Thanks to Jeffrey Sofferin
349 <sofferin@users.sourceforge.net> for reporting this bug.
350
351 6. Fixed a bug which caused restore, in some particular cases,
352 to ask some 'scary' questions and leave a bunch of RSTTMP
353 directories behind when restoring incremental tapes. Thanks
354 to Philippe Troin <phil@fifi.org> for reporting this bug and
355 providing the test cases.
356
357 7. Changed the wording when inodes are excluded from dump:
358 replaced 'Added inode 7 to exclude list' with
359 'Excluding inode 7 (resize inode) from dump', as suggested
360 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
361
362 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
363 ===================================================================
364
365 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
366 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
367 <eros@lamel.bo.cnr.it> for reporting the bug.
368
369 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
370 ===================================================================
371
372 1. Made dump print out the ext2 logical block number in case of a read
373 error, which can be given as an argument to debugfs' ncheck command,
374 in order to find out the affected inode. Added note about this
375 usage in dump's man page.
376
377 2. Fixed a problem in restore when reading tapes written on big
378 endian machines with very old versions of dump. The patch was
379 contributed by George Helffrich <george@geology.bristol.ac.uk>.
380
381 3. Fixed the tape length calculation when using large tapes
382 and compression. Thanks to Georg Lippold
383 <g_lippold@sourceforge.net> for reporting the bug.
384
385 4. Added a new set of examples in dump_on_cd_2 directory, based
386 on dump_on_cd examples but somewhat enhanced, supporting DVD
387 media, and localized in english and german. Thanks to
388 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
389
390 5. Made dump save 32 bit UID/GID. Internally, this was achieved
391 by switching from the old BSD inode format to the new BSD
392 inode format, which means that the tape format was changed.
393 However, since all restore versions out there should
394 transparently support both inode formats, the change should
395 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
396 reporting the bug.
397
398 6. Fixed a lot of warnings in the code shown when compiling
399 with 'gcc -W'. Thanks to Matthias Andree
400 <matthias.andree@stud.uni-dortmund.de> for reporting this.
401
402 7. Fixed a small markup bug in the dump man page. Thanks to
403 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
404 patch.
405
406 8. Rewrote entirely the man pages using the tmac.an macro
407 package (Linux man page format) instead of the original BSD
408 format. They should be now cleaner and easier to modify.
409
410 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
411 ==================================================================
412
413 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
414 non Linux clients were misinterpreted. The description of the problem
415 (incompatible numbering in Linux mtio opcodes) is documented at
416 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
417 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
418 providing an excellent, cross-platform replacement for rmt in his
419 star package.
420
421 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
422 which caused restore to fail when the inode maps were not entirely
423 contained on the first volume (when using really small volumes or
424 when dumping a huge number of inodes).
425
426 3. Fixed a problem in dump, where files in subdirectories of directories
427 that have the nodump flag set (or being excluded with -e/-E) were
428 sometimes incorrectly dumped (instead of being excluded). The original
429 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
430 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
431 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
432 help testing the patch.
433
434 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
435 which makes one able to pipe the output of dump, by the net, to
436 a remote CD-burner server.
437
438 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
439 and use GNU's symbolic syntax over rmt instead of numerical values
440 to assure multiple platform compatibility.
441
442 6. Documented the -d option in restore.
443
444 7. Added a -v (verbose) mode to dump. For now it just prints the number
445 of the inode being dumped, but this could evolve in future versions
446 to include interesting debugging output.
447
448 8. Added a -o flag to restore, which automatically restores the current
449 directory permissions in -i or -x mode, without asking the operator
450 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
451 and Peter Samuel <peters@e-smith.com>.
452
453 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
454 ====================================================================
455
456 1. Fixed a bug in the QFA feature which could have caused
457 a bad tape positionning on restore (causing extra delays in
458 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
459
460 2. Added a small note in the dump man page specifying that
461 there should be no white space between the option letter and
462 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
463
464 3. Made restore work with older versions of the readline library,
465 by checking in configure for several extended symbols. Restore
466 can now be compiled with a readline as old as the 2.0 release
467 (though it may be a good idea to upgrade it to a more recent
468 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
469 reporting the build failures.
470
471 4. Fixed a performance problem with the QFA file creation in
472 dump, which made unnecessary seeks on the tape slowing down
473 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
474 reporting this issue.
475
476 5. Removed the inclusion of some kernel headers in the dump
477 source, which prevented the compile in some kernel/glibc
478 headers/architecture combination. Thanks to Bdale Garbee
479 <bdale@gag.com> for reporting the bug.
480
481 6. Added the appropriate error message when dump fails to
482 open the output file for writing. Thanks to Amith Varghese
483 <amithv@yahoo.com> for reporting this bug.
484
485 7. Made restore able to understand large Solaris ufsdump tapes
486 (containing inodes bigger than 4194304). Sun have introduced
487 an "extension" to the dump tape format when dealing with
488 those inodes, which was uncorrectly handled by Linux restore.
489 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
490 providing a test case.
491
492 8. Added the -m parameter to dump which optimises the output for
493 inodes having been changed but not modified since the last dump
494 ('changed' and 'modified' have the meaning defined in stat(2)).
495 For those inodes, dump will save only the metadata, instead of
496 saving the entire inode contents. Inodes which are either
497 directories or have been modified since the last dump are saved
498 in a regular way. Uses of this flag must be consistent, meaning
499 that either every dump in an incremental dump set have the flag,
500 or no one has it.
501
502 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
503 =======================================================================
504
505 1. Fixed behaviour of dump when exceeding resource limits
506 (SIGXFSZ treatment).
507
508 2. Added the -L flag to restore to allow the user to specify a
509 maximal allowed number of miscompares when using restore
510 with the -C option to check the backup.
511
512 3. Detailed the manual entry for the -N option of restore.
513
514 4. Added the -a flag to restore to make able doing unattended
515 restores in -i or -x mode (automatically walks through the
516 multiple dump volumes).
517
518 5. Extended the QFA mode to work with local files and/or
519 remote tapes and files. This way, restore can know in advance
520 the tape number and the offset for the inodes to extract and
521 can minimize the extraction time by seeking directly to the
522 good tape/offset.
523
524 6. Added the -A <archive> option to both dump and restore,
525 which makes dump to archive a dump table-of-contents in
526 the specified file to be used by restore to determine
527 whether a file is in the dump file that is being restored.
528 (the archive file syntax is also compatible with the
529 Solaris ufsdump generated one).
530
531 7. Small fix in restore making it able to read some (broken ?)
532 Solaris ufsdump tapes.
533
534 8. Fixed dump to correctly recognise the root filesystem when using
535 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
536 for reporting this bug.
537
538 9. Added the -P <file> option to restore to create a
539 Quick File Access file from an already made dump. Patch
540 contributed by Uwe Gohlke <uwe@ugsoft.de>.
541
542 10. Made restore compile and run on Solaris, making it a
543 possible replacement for the standard ufsrestore. Port was
544 contributed by Uwe Gohlke <uwe@ugsoft.de>.
545
546 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
547 =====================================================================
548
549 1. Added a set of backup scripts from Eugenio Diaz
550 <getnito@yahoo.com> in the examples section. It features
551 automatic (cron based) full and incremental dumping of
552 several filesystems on a separate filesystem.
553
554 2. Fixed a off-by-one miscalculation which disabled dumping
555 a one letter subdirectory. Thanks to Chris Siebenmann
556 <cks@utcc.utoronto.ca> for reporting the bug.
557
558 3. Fixed several restore bugs occuring when trying to
559 restore some missing files on the tape. Thanks to Chris
560 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
561
562 4. Fixed --with-ldopts configure argument passing, installing from
563 a separate object directory, makefile cleanups contributed
564 by <splite@purdue.edu>.
565
566 5. Fix a bug which could caused, in some conditions, the highest
567 number inode of a filesystem, to not be dumped. Many thanks
568 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
569 chase this bug.
570
571 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
572 =======================================================================
573
574 1. Added a mini howto from Patrick Walsh in the examples
575 directory.
576
577 2. Minor man pages syntax corrections. Thanks to
578 Chris Danis <screechco@home.com> for reporting the bugs.
579
580 3. Added a script from David B. Peterson <dave@toppledwagon.com>
581 to the examples section. It features dumping several
582 filesystems to a remote tape drive upon ssh.
583
584 4. Added a patch provided by Richard Jones <rich@annexia.org>
585 which allows BRADEMAX (number of read errors tolerated by
586 dump) to be adjusted using the -I option.
587
588 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
589 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
590 patch.
591
592 6. Add the -l option to restore to specify if, when doing a
593 remote restore, the file used to access the backup is a
594 regular file (the defaults being a tape drive). Restore needs
595 to know this information when reading a remote compressed
596 dump. Previously, this information was autodetected, but
597 the autodetection code fails (with ioctl: Inappropriate ioctl
598 for device) when using a non Linux remote box. Thanks to
599 many users and especially to Eros Albertazzi
600 <eros@lamel.bo.cnr.it> for reporting this.
601
602 7. Found a workaroung for the dump deadlock problem (3 childs
603 stuck in pause(), father in read()). The workaround seems
604 to work for me and several beta-testers. If it doesn't work
605 for you, please report back.
606
607 8. Updated the RPM spec file (BuildPrereq, URL etc).
608
609 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
610 ========================================================================
611
612 1. Fixed the permissions of a newly created QFA file by dump.
613
614 2. Cleaned up the source of dump (the external variables
615 definition was a complete mess, making possible to have
616 objects overlap).
617
618 3. Fixed restore to use the full tape volume path when doing
619 a compare (since it changes the working directory to the
620 filesystem being compared in the process).
621
622 4. Added the -q option to dump which makes dump abort
623 whenever it needs operator attention. It should be
624 easier to use dump in scripts with this option.
625
626 5. Detect the use of incompatible options to dump and
627 refuse them (like -a and -B options together).
628
629 6. Added bzip2 compression to dump/restore (use option -j level
630 to select it). Note that you will need a restore version
631 >= 0.4b24 in order to restore a bzip2 compressed dump.
632 The same warning as for the zlib compression applies:
633 the tape format of a bzip2 dump is not compatible with the
634 original BSD tape format.
635
636 7. Fixed a overflow problem in dump corrupting the dump when
637 very large files were encountered. Thanks to Vladimir Nadvornik
638 for the bug report.
639
640 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
641 kernel buffer/page cache before starting the dump, helping
642 a bit those who use dump on mounted filesystems. Thanks to
643 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
644 for this suggestion.
645
646 9. Updated the RPM spec file following the RedHat changes
647 (dynamically linked binaries now in /usr/sbin etc).
648
649 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
650 which enables restore to recognise multi volume compressed dumps
651 done on CD. Included his scripts for dump (which pipe the dump
652 output directly into cdrecord) and restore. There is now
653 possible to to backups to CD on the fly!
654
655 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
656 ===================================================================
657
658 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
659 Marcus Meissner (Caldera International Security Dept.).
660
661 2. Implement the Sun rmt extended protocol. Patch submitted
662 by Ian Gordon <iangordon@users.sourceforge.net>.
663
664 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
665 the linux kernel header. This ensures that dump always has the
666 latest version of this file. Patch submitted by
667 Andreas Dilger <adilger@turbolinux.com>.
668
669 4. Report any filesystem present in either /etc/fstab with a
670 positive passno or /etc/dumpdates in dump -w output.
671 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
672
673 5. Fixed the looping problem in dump introduced in the
674 previous version.
675
676 6. Changed the -B option of dump to limit the size of
677 _compressed_ output per volume if compression is on.
678 Patch contributed by Helmut Jarausch
679 <jarausch@igpm.rwth-aachen.de>. Note however that, since
680 it is impossible to predict the size of the compressed
681 data before doing the compression, the -B limit is a bit
682 conservative.
683
684 7. Fixed a bug in reading the operator typed file/tape path for
685 the next volume in restore.
686
687 8. Implemented a "-F script" option to restore which permits the
688 user to specify a script which will be launched at the
689 beginning of each tape, useful for automatic programming of
690 tape changers for example. See the restore man page for the
691 script parameters and return codes.
692
693 9. Small fix for the QFA routines provided by Uwe Gohlke
694 <uwe@ugsoft.de>, and some recommendations for QFA uses in
695 the man pages.
696
697 10. Fixed the multivolume restoring where making a mistake
698 to the 'Mount next tape' prompt caused several blocks to
699 be lost.
700
701 11. Enhanced the -e option of dump to take as a parameter a
702 comma separated list of inode numbers.
703
704 12. Added the -E option to dump which specify a file containing
705 inode numbers to exclude from the dump.
706
707 13. Fixed the compressed multi-volume dump + restore.
708
709 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
710 ==================================================================
711
712 1. Made dump report the number of blocks dumped per volume.
713 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
714
715 2. Fix a bug in dump where waiting too much at the 'change volumes'
716 question caused the volume to be rewritten. Thanks to
717 Graham Reed <greed@users.sourceforge.net> for reporting the
718 bug and providing a patch.
719
720 3. Added a compression option to dump, contributed by Jerry
721 Peters <gapeters@worldnet.att.net>.
722
723 WARNING: the compression works only on files/pipes and
724 drives supporting variable block size.
725
726 WARNING2: tapes created using the compression feature are
727 incompatible with the standard dump tape format, and a
728 version of restore >= 0.4b22 will be needed for extraction.
729
730 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
731 architectures. Thanks to Paul Slootman <paul@debian.org> for
732 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
733 it upstream.
734
735 5. Many cleanups (CPP defines, const char warnings, check of
736 ext2fs COMPAT flags, time_t cleanups, added libext2 version
737 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
738
739 6. Made --prefix option work in configure. All the install path
740 are now based on the configure parameters.
741
742 7. Added the Quick File Access mode in dump/restore, contributed
743 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
744 a file tape position for each inode, and this file is used by
745 restore (if called with parameter Q and the filename)
746 to directly position the tape at the file restore is currently
747 working on. This saves hours when restoring single files from
748 large backups, saves the tapes and the drive's head. Use
749 --enable-qfa option of configure to compile in the QFA support.
750
751 8. Added the possibility to dump several files and directories
752 in a single invocation of dump. Thanks to Uwe Gohlke
753 <uwe@ugsoft.de> for implementing this option.
754
755 9. Fixed the dumping and restoring of files with holes
756 (files ending with holes didn't get dumped or restored
757 correctly).
758
759 10. Fixed a socket descriptor leak in dump, which leaved opened
760 3 file descriptors per dump process (and there is one dump
761 process per tape).
762
763 11. Fixed dump large file system support, by explicit use of
764 open64/lseek64/etc functions (explicit use needed because
765 e2fsprogs libraries don't behave well when compiled with
766 FILE_OFFSET_BITS=64).
767
768 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
769 ======================================================================
770
771 1. Fixed some bugs in the dump -w|-W logic introduced by
772 the previous version. Thanks to Andreas Dilger
773 <adilger@turbolinux.com> for his help on this one.
774
775 2. Fixed again a compilation problem when using old e2fs
776 headers (filesystem label related). Thanks to many users
777 who reported this stupid error.
778
779 3. Fixed a build problem on old lib5 based systems dealing with
780 _PATH_MNTTAB being undefined. Thanks to John Adams
781 <johna@onevista.com> for reporting the bug.
782
783 4. Improved the error detection in restore when creating
784 the temporary files in TMPDIR. Now restore will corectly
785 report a 'no space left on device' error instead of
786 strange errors which could imply an archive corruption.
787 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
788 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
789
790 5. Added the throughput information in dump's progression
791 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
792 for the patch.
793
794 6. Use libext2fs's inode scanning routines, which are particularly
795 robust when dealing with errors (especially when having some
796 bad blocks on the filesystem). Many thanks to Colin
797 <colin@horizon.com> for reporting the bug, investigating
798 the issues, finding a workaround, writing the patch and
799 fully testing it... (of course, if this patch will break
800 anything, he is to blame for :-)).
801
802 7. Made dump and restore LFS aware. Dump can dump a filesystem
803 containing large files, generate a large file on output and
804 restore can restore them. This feature is not enabled by
805 default, use --enable-largefile option of configure to enable
806 it (you need to have a LFS aware glibc though). Thanks to
807 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
808 and to Theodore T'so <tytso@valinux.com> for his always
809 useful thoughts.
810
811 8. Made dump ask upon a tape write error if it should rewrite
812 the current volume (assume this is a bad tape) or if it should
813 assume an end-of-tape condition (useful for tape drives which
814 behaves badly at the end of the tape). Thanks to Andreas
815 Dilger <adilger@turbolinux.com> for the suggestion.
816
817 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
818 =======================================================================
819
820 1. Fixed a small compilation problem due to a change
821 in the definintion of the struct sigaction in
822 glibc 2.0 / libc5. Thanks to Gunther Schlegel
823 <schlegel@riege.de> for reporting the bug and to
824 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
825 a fix.
826
827 2. Modified the label and uuid analysis in order to be
828 self-contained (independant of kernel/libc headers). This
829 should fix the compile with older kernel/libc headers and
830 will preserve the functionality. Thanks to Bernhard Erdmann
831 <bernhard.erdmann@gmx.de> for reporting the bug.
832
833 3. The 'exclude inode' option, if applied to a directory
834 inode, excludes now correctly all the leaves of this
835 directory. Thanks to John R. Dennison
836 <gerdesas@users.sourceforge.net> for reporting the bug.
837
838 4. Fixed the '-e' option to disable the possibility
839 to exclude the root inode (which causes the dumps to
840 be unreadable by restore). Prevented array overflow
841 when multiple -e options are used.
842
843 5. Fixed dump to correctly interpret a filesystem argument
844 which is a mountpoint and it is not an absolute path
845 (as specified in the fstab). Thanks to Bernhard R. Erdmann
846 <be@berdmann.de> for reporting the bug.
847
848 6. Made dump able to backup files larger than 2 GB. Note that
849 dump still doesn't cope with files larger than 4 GB.
850
851 7. Restore the real uid and gid before invoking an external
852 RSH program (big hole when dump or restore is suid root!).
853
854 8. Get the values for the default tape device and the location
855 of fstab file from the system headers. Thanks to
856 Andreas Dilger <adilger@turbolinux.com> for the patch.
857
858 9. Made dump -w|-W report all recognized filesystems
859 present in either /etc/fstab or /etc/dumpdates, and present
860 the list in the same order as in fstab file. Thanks
861 to Andreas Dilger <adilger@turbolinux.com> for the patch.
862
863 10. Made dump's -a (automatic end of tape detection) the
864 default. Specifying one of -B, -c, -d or -s options will
865 override the EOM detection. Thanks to Andreas Dilger
866 <adilger@turbolinux.com> for the patch.
867
868 11. Save the ext2 filesystem label into the dump volume label.
869 Specifying a volume label on the command line (-L option)
870 will override this feature. Thanks to Andreas Dilger
871 <adilger@turbolinux.com> for the patch.
872
873 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
874 =====================================================================
875
876 1. Fixed the signal handling in dump (which I broke in 0.4b17)
877 which was causing several strange dump failures (dump
878 hanged or segmentation faults etc).
879
880 2. Specified the default blocksize in the dump man page.
881
882 3. Changed two info messages of restore to be written on stdout
883 instead of stderr in order to leave stderr only for errors
884 or warnings. Thanks to Stephen Atwell
885 <satwell@urbana.css.mot.com> for the suggestion.
886
887 4. Corrected an off by one calculation which prevented
888 dumping a directory having a 1 character name length.
889 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
890 for reporting the bug.
891
892 5. Reinforce dump in order to allow only level 0 and no
893 -u option when dumping a subdirectory, and document
894 this in the man page. Thanks to Bernhard Erdmann
895 <bernhard.erdmann@gmx.de> for reporting the bug.
896
897 6. Fixed a small harmless bug in restore which caused
898 in some conditions hard links to be restored several
899 times (therefore generation some warning messages).
900 Thanks to Kenneth Porter <shiva@well.com> for
901 reporting the bug.
902
903 7. Updated the RPM spec file to the latest RedHat version,
904 providing FHS packaging and other cosmetic changes.
905 You will need RPM version >= 3.0.5 in order to build the RPM.
906
907 8. Updated the configure script to check for libtermcap
908 before checking for libreadline, because we need this
909 library in order to compile the readline support.
910
911 9. Made dump understand the LABEL= and UUID= notation
912 both in /etc/fstab and on the command line. Note that
913 you will need the /proc filesystem in order to use
914 these notations. Thanks to Erik Troan <ewt@redhat.com>
915 for providing the patch.
916
917 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
918 ===================================================================
919
920 1. Fixed a potential buffer overflow in restore. Thanks
921 to Stan Bubrouski <satan@fastdial.net> for reporting
922 the bug.
923
924 2. Fixed a readline-related bug which prevented
925 'cat DUMPFILE | restore -i -f -' from working. Thanks
926 to Charles Karney <karney@users.sourceforge.net>
927 for the bug report.
928
929 3. Changed a few "panic" into "exit", causing restore to
930 be more stable against some attacks (like the last one
931 reported on Bugtraq, although the last version of restore
932 was not vulnerable - just dumped core). Thanks to
933 Andreas Hasenack <andreas@conectiva.com.br> for reporting
934 the bugs to me.
935
936 4. Removed the suid-root bit on dump and restore in the
937 default build (and generated RPMs). It should be safer
938 now. Users who need the suid-root capabilities in order
939 to make network backups should read first the man page
940 of dump and enable the suid bit manually.
941
942 5. Added -ltermcap to the compile parameters for restore
943 when using readline support, in order to make the compile
944 process work on some systems (SuSE ?). Thanks to
945 Patrik Schindler <poc@pocnet.net> for reporting the bug.
946
947 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
948 ====================================================================
949
950 1. The -F script is called now *only* at the end of a tape,
951 not at the end of the dump. Thanks to Christian Haul
952 <haul@informatik.tu-darmstadt.de> for the bug report.
953
954 Normally, the device name and the current volume number
955 are passed on the command line of the script. However,
956 if you want the old style script (with no arguments
957 passed on the command line), you can enable it in
958 configure with the --enable-oldstylefscript.
959
960 2. Use posix signal handling to preserve dump functionality
961 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
962 the patch.
963
964 3. Made the exit code of restore in case of a 'C'ompare
965 command reflect the result of the compare. An exit status
966 of 0 means the dump archive is correct, 1 means tape errors,
967 2 means that some files were modified. Thanks to Kenneth Porter
968 <shiva@well.com> for the suggestion.
969
970 4. Made (finally) quotes work in interactive restore.
971
972 5. Small fixes in order to allow dump to compile with a
973 really old e2fsprogs version. Thanks to Ian Zimmerman
974 <itz@speakeasy.org> for the bug report.
975
976 6. Add GNU readline capabilities to interactive restore.
977 Use configure's --enable-readline flag to enable this feature.
978 Thanks to Patrik Schindler <poc@pocnet.net> for the
979 suggestion.
980
981 7. Do the compare on the fly in restore 'C' mode (this will
982 allow not to exhaust the available /tmp space when
983 ccmparing large files). Thanks to Kenneth Porter
984 <shiva@well.com> for the suggestion.
985
986 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
987 ====================================================================
988
989 1. Fixed some several duplicate 'DUMP: DUMP:' in the
990 output of dump.
991
992 2. Corrected the estimation of blocks to dump. Note that
993 this is still wrong for large directory inodes, due
994 to the size of a BSD directory entry on the tape
995 (estimation is based on the size of the ext2 directory,
996 which is smaller because it doesn't contain the null
997 character at the end).
998
999 3. Made dump report the total number of MB written to
1000 tape at the end of run. Thanks to W. Reilly Cooley
1001 <wcooley@nakedape.cc> for the patch.
1002
1003 4. Added the -X option to restore giving the possibility
1004 to read the names of the files to be extracted/listed
1005 from a text file (in addition of the command line).
1006 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
1007 patch.
1008
1009 5. Added the device name and the current volume number
1010 as arguments to the end of tape script (dump -F option).
1011
1012 6. Made the multi-volume dump work again (after having
1013 broken it in 0.4b15).
1014
1015 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
1016 ===================================================================
1017
1018 1. Added a prompt command in interactive restore mode. Thanks
1019 to Andreas Dilger <adilger@home.com> for the patch.
1020
1021 2. Fixed a buffer overflow problem in dump (caused by
1022 not checking the size of the filesystem parameter).
1023 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
1024 reporting this on Bugtraq (and to several dump users
1025 who forwarded me his mail).
1026
1027 3. Added the '-F script' option to dump in order to
1028 launch a script at the end of each tape (to be used
1029 with a tape changer, or to notify the sysadmin by
1030 pager etc.).
1031
1032 4. Fixed a bug in restore compare code caused by the changes
1033 I made in 0.4b14.
1034
1035 5. Fixed the treatment of options using the old BSD syntax
1036 in both dump and restore.
1037
1038 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
1039 =======================================================================
1040
1041 1. Fixed a bug in dump which may have caused invalid deleted
1042 directories to be dumped out if they were deleted after the
1043 mapping pass. This could occure on active filesystem and lead
1044 to heap corruption (causing dump malfunction in many possible ways).
1045 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
1046 bug and submitting the patch.
1047
1048 2. Fixed the handling of the filesystem parameter in dump. A
1049 '/mnt/disk' parameter caused the disk contents to be dumped,
1050 but a '/mnt/disk/' parameter caused the mountpoint directory
1051 to be dumped (generally an empty directory).
1052
1053 3. Improved the output of dump in order to tell which directory
1054 it is currently dumping (when dumping a subtree).
1055
1056 4. Added the '-e' exclude inode option to dump. Thanks to
1057 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
1058
1059 5. Added a REPORTING-BUGS file in order to provide a guide
1060 on how to correctly report a bug in dump/restore.
1061
1062 6. Fixed a restore problem when restoring a hard link to an inode
1063 having the immutable or append-only attribute set. Thanks to
1064 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
1065
1066 7. Fixed a compatibility problem between dumps made on little
1067 endian machines (the format was correct) and big endian
1068 machines (the format was incorrect). This fix break the
1069 compatibility with the older dumps made on big endian
1070 machines (sparc, m86k, ppc etc). For the first time in
1071 linux dump's history (I believe), the dumps made by *BSD,
1072 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
1073 so interchangeable. Thanks to Rob Cermak
1074 <cermak@ahab.rutgers.edu> for submitting the bug and
1075 helping me test the fix.
1076
1077 8. Fixed the way dump reports the remaining percent/time, if
1078 the number of blocks actually dumped exceeds the estimated
1079 number of blocks. Thanks to Jean-Paul van der Jagt
1080 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
1081
1082 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
1083 ======================================================================
1084
1085 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
1086 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
1087
1088 2. Removed the requirement to build the RPM as root from the
1089 spec file. Thanks to Christian Weisgerber
1090 <naddy@mips.rhein-neckar.de> for submitting this.
1091
1092 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
1093 causing dump to block when dumping really huge filesystems.
1094 Many thanks to Patrik Schindler <poc@pocnet.net> for
1095 helping me find this bug.
1096
1097 4. Fixed the treatment for an interrupt signal when dump access
1098 the remote tape through RSH. Thanks to Christian Weisgerber
1099 <naddy@mips.rhein-neckar.de> for providing the patch.
1100
1101 5. Fixed a bug which was causing dump/restore to display
1102 garbage characters instead of the remote host name.
1103
1104 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
1105 =====================================================================
1106
1107 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
1108 <kukuk@suse.de> for submitting the patch.
1109
1110 2. Fix for the exit code when using the size estimate option of
1111 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
1112 submitting the patch.
1113
1114 3. Handle EINTR in atomical reads/writes in dump, which was causing
1115 dump to fail on some systems. Thanks to Eric Jergensen
1116 <eric@dvns.com> for reporting the bug and submitting the patch.
1117
1118 4. Handle more than 16 characters for the device names in dumpdates.
1119 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
1120 tracking down the problem and proposing the solution.
1121
1122 5. Fixed a bug in dump which prevented the creation of the
1123 dumpdates file when doing a 0-level dump without already
1124 having a dumpdates file. Thanks to Patrik Schindler
1125 <poc@pocnet.net> for reporting the bug.
1126
1127 6. Changed the way dump 'S' flag reports the size estimate
1128 from number of blocks into bytes (making it compatible
1129 with the Solaris version, and simplifying things for
1130 amanda users). Thanks to Jason L Tibbitts III
1131 <tibbs@math.uh.edu> for reporting the bug.
1132
1133 7. Fixed a compatibility problem in linux/alpha dump tape format.
1134 Now the linux/alpha dump are (again) compatible with the
1135 other dump formats. But this breaks compatibility with
1136 older dumps made on alpha. Thanks to Mike Tibor
1137 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
1138
1139 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
1140 ======================================================================
1141
1142 1. Added a '--enable-kerberos' to configure.
1143
1144 2. Added a 'S' option to dump which determines the amount of space
1145 that is needed to perform the dump without actually doing it, similar
1146 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
1147 <cermak@ahab.rutgers.edu>.
1148
1149 3. Added a 'M' multi-volume option to dump and restore which enables
1150 dumping to multiple files (useful when dumping to an ext2
1151 partition to make several dump files in order to bypass the 2GB
1152 file size limitation). The 'f' argument is treated as a prefix and
1153 the output files will be named <prefix>001, <prefix>002 etc. With
1154 the 'M' flag, restore automatically selects the right file without
1155 asking to enter a new tape each time.
1156
1157 4. Fixed a memory leak which was causing dump to grow very big
1158 (270MB when dumping a 10GB filesystem...). Thanks to Jason
1159 Fearon <jasonf@netrider.org.au> for reporting the bug.
1160
1161 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
1162 ======================================================================
1163
1164 1. Make configure test if the system glob routines support
1165 extended syntax (ALTDIRFUNC). If not, use the internal glob
1166 routines instead of system ones. Thanks to Bernhard Sadlowski
1167 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
1168 and helping me resolve this and other minor libc5 compiling
1169 glitches.
1170
1171 2. Fix a problem when dumping a ext2fs with the 'filetype'
1172 feature enabled. Thanks to Patrick J. LoPresti
1173 <patl@cag.lcs.mit.edu> for reporting the bug and to
1174 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
1175
1176 3. Made the nodump flag work on directories. A directory which
1177 has the nodump flag gets never dumped, regardless of its
1178 contents.
1179
1180 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
1181 which allows dump on an active ext3 filesystem. However, this
1182 is a "quick and dirty" patch which enables backup of an ext3
1183 filesystem through the ext2 compatibility (by ignoring the
1184 NEEDS_RECOVERY bit). The journal file is not recognized and
1185 it is dumped (it should not).
1186
1187 5. Test the superblock compatibility flags when dumping, in order
1188 to be sure that we know how to deal with specific features.
1189
1190 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
1191 ====================================================================
1192
1193 1. Use lchown instead of chown, fixing a possible security problem
1194 when restoring symlinks (a malicious user could use this
1195 to deliberately corrupt the ownership of important system files).
1196 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
1197 this and providing the patch.
1198
1199 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
1200 ====================================================================
1201
1202 1. Put dump sources under CVS, added Id tags in all files so
1203 one can use 'ident' on binary files.
1204
1205 2. Added the dump/restore version in the usage text so one can
1206 easily verify the version he is using.
1207
1208 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
1209 a va_start/va_end problem on linux-ppc (always call va_start
1210 va_end in pairs each time we use a vararg function).
1211
1212 4. Added again the DT_* constants because old libc does not
1213 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
1214 for submitting the bug report.
1215
1216 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
1217 this should enable dumping big (huge) filesystems.
1218
1219 6. Added the RSH environment variable in order to be able to
1220 use a rsh replacement like ssh when doing remote backups (and
1221 bypass the security limitations of rcmd). Now you can do remote
1222 backups without being root (or making dump setuid root).
1223
1224 7. Modified again the way dumpdates works. For incremental dumps,
1225 we need to read dumpdates even if we are not using 'u' option.
1226 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
1227 this should work.
1228
1229 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
1230 ===================================================================
1231
1232 1. Removed the 'k' flag from the restore 'about' text if kerberos
1233 was not compiled in.
1234
1235 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
1236 to them (fsetflags takes a char*, setflags an open fd!).
1237
1238 3. (f)setflags is called only if the flags aren't empty. If the
1239 file is a special file, a warning is printed, because changing
1240 flags implies opening the device. Normally, a special file
1241 should not have any flag... (Debian bug #29775, patch provided
1242 by Abhijit Dasgupta <abhijit@ans.net>).
1243
1244 4. Made possible to dump a file system not mentioned in /etc/fstab.
1245 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
1246
1247 5. Changed the default behaviour to not create dumpdates
1248 unless 'u' option is specified. Removed the old "debian-patch"
1249 which provided the same thing. (Debian bug #38136, #33818).
1250
1251 6. Removed all those dump*announce, since they were getting old...
1252
1253 7. Added warning messages if dumpdates does not exist and
1254 when an inferior level dump does not exist (except for a level 0
1255 dump).
1256
1257 8. Debugged the glob calls in interactive mode: restore used a
1258 dirent struct which was different from the /usr/include/dirent.h
1259 one (this used to work, is it a glibc2 change?), so none of the
1260 compat glob (which used /usr/include/dirent.h) or the system glob
1261 worked. Restore use now the system dirent (and the system
1262 DT_* constants), which are compatible with BSD ones.
1263
1264 9. Added a configure flag (--with-dumpdatespath) to specify
1265 the location of dumpdates file. By default, it is
1266 /etc/dumpdates.
1267
1268 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1269 included the current date/version in man pages.
1270
1271 11. Corrected the estimation of remaining time when
1272 the operator doesn't change the tapes quickly enough. This
1273 was an old bug, I thought I corrected it, and discovered
1274 that in fact it was corrected in two different places, so
1275 the results canceled each other...
1276
1277 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1278 ===================================================================
1279
1280 1. Integrated multiple patches from RedHat, Debian and SuSE:
1281
1282 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1283 isn't specified.
1284 - several fixes in the man pages.
1285 - update the default tape device to /dev/st0.
1286 - many updates for Linux Alpha (byte ordering, size_t etc).
1287 - buffer overruns.
1288 - use environment variable for TMPDIR (instead of /tmp).
1289 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1290 - workaround egcs bug (RedHat bugs #4281 and #2989).
1291 - wire $(OPT) throughout Makefile's.
1292
1293 2. Upgrade the dump revision to 1, making possible to dump filesystems
1294 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1295
1296 3. Fix some compile warnings, prototype all functions.
1297
1298 4. Use glibc err/glob instead of internal compatibility
1299 routines (only if available).
1300
1301 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1302 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1303
1304 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1305 ======================================================================
1306
1307 1. Integrated the changes from FreeBSD-3.1-RELEASE
1308 (mostly bug fixes, buffer overruns, dump has now an "automatic
1309 tape length calculation" flag, dump/restore can use kerberos now
1310 (this is NOT tested), use environment variables for TAPE and
1311 RMT etc.).
1312
1313 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1314
1315 3. Corrected a bug in restore when using 'C' option with multi-volumes
1316 tapes (files splited accros two tapes give "size changed" errors
1317 when comparing).
1318
1319 4. Corrected the long standing bug when dumping multiple tapes.
1320 This works for me, needs further testing.
1321
1322 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1323 ====================================================================
1324
1325 1. Dump now runs correctly on kernels 2.1.x
1326 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1327
1328 Changes between versions 0.4b2 and 0.4b3
1329 ========================================
1330
1331 1. Use realpath() if available
1332
1333 2. Report statistics
1334
1335 Changes between versions 0.4b1 and 0.4b2
1336 ========================================
1337
1338 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1339 the patch)
1340
1341 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1342
1343 3. Fixed configure.in to correctly handle --enable-rmt
1344
1345 Changes between versions 0.3 and 0.4b1
1346 ======================================
1347
1348 1. Integrated the changes from 4.4BSD-Lite2
1349
1350 2. Integrated the patches from Debian and Red Hat
1351
1352 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1353
1354 4. Changed dump to use the Ext2fs library to get block addresses. This
1355 should solve the endianness problem on SparcLinux.
1356
1357 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1358 distribution's one) to use autoconf
1359
1360 6. Fixed a few minor bugs
1361
1362 Changes between versions 0.2e and 0.2f
1363 ======================================
1364
1365 1. Added the creation of named pipes (fifos) in restore.
1366
1367 2. Added the -N flag in the restore manual page.
1368
1369 3. Added the file kernel.patch which contains the llseek() optimization
1370 patch for 1.2.x kernels.
1371
1372 4. Fixed a bug in the restoration of symbolic links: owner and group were
1373 not restored.
1374
1375 5. Integrated some changes from FreeBSD 2.2.
1376
1377 6. Added a call to ftruncate() after restoring each file to restore
1378 correctly files ending by a hole.
1379
1380 Changes between versions 0.2d and 0.2e
1381 ======================================
1382
1383 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1384 with owner = root (0) and group = root/wheel/whatever (0).
1385
1386 Changes between versions 0.2c and 0.2d
1387 ======================================
1388
1389 1. Dump is now able to backup 2GB+ filesystems.
1390
1391 2. Dump and restore can now be linked as static binaries.
1392
1393 Changes between versions 0.2b and 0.2c
1394 ======================================
1395
1396 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1397
1398 Changes between versions 0.2a and 0.2b
1399 ======================================
1400
1401 1. Really fixed the bug that I should have corrected in 0.2a.
1402
1403 2. Enabled optimization again.
1404
1405 Changes between versions 0.2 and 0.2a
1406 =====================================
1407
1408 1. Disabled the optimization during compilation.
1409
1410 Changes between versions 0.1 and 0.2
1411 ====================================
1412
1413 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1414 the fs_type field (actually, I modified the file fstab.c to make it
1415 use the mntent functions).
1416
1417 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1418 made by dump should be readable by the BSD restore and backups made
1419 by the BSD dump should be readable by restore. Unfortunately, this
1420 means that the dump format has changed between version 0.1 and version
1421 0.2 :-(
1422
1423 3. Dump is now able to backup a subtree, it is no longer limited to whole
1424 filesystems like the BSD version.
1425
1426 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1427 than 2 GB.
1428
1429 Changes between versions 0.0 and 0.1
1430 ====================================
1431
1432 1. Now create links rdump and rrestore during the `make install' step.
1433
1434 2. Linux port specific bugs added to the manual pages
1435
1436 3. Incorrect estimation of the number of tapes blocks fixed when doing
1437 incremental backups.
1438
1439 4. Better ls-like format in restore in interactive mode.