Test for libncurses, then libtermcap when --enable-readline'ing
[dump.git] / CHANGES
1 $Id: CHANGES,v 1.256 2004/07/20 15:52:58 stelian Exp $
2
3 Changes between versions 0.4b37 and 0.4b38 (released ????????????)
4 ==================================================================
5
6 1. Fix a couple of troff syntax bugs in the man pages.
7 Thanks to Eric Raymond <esr@thyrsus.com> for the patch.
8
9 2. Made restore use either libncurses or libtermcap, depending
10 on which one is available at configure time.
11
12 Changes between versions 0.4b36 and 0.4b37 (released July 7, 2004)
13 ==================================================================
14
15 1. Added the --enable-staticz configure option which enables
16 dump and restore to be linked with static versions of
17 libz and libbz2 (and dynamic versions of all other libraries).
18 This will make Debian users happy, because libz and libbz2
19 were the only needed libraries living in /usr, all the
20 others live in /lib. In case of system emergency, it is
21 better not to have to rely on an extra filesystem.
22 Thanks to Bdale Garbee <bdale@gag.com> for the suggestion.
23
24 2. Fix compilation on (at least the Linux Debian port to) AMD64.
25 (<ext2fs/ext2_types.h> defines some types (__s64 and __u64)
26 that are also defined by <linux/types.h> (<asm/types.h>) and
27 they conflict).
28
29 3. Make dump's reading of the dumpdates file a bit more robust,
30 preventing dump from crashing when the dumpdates file has
31 been modified by hand.
32
33 4. Fixed some offset calculations in dump code which could
34 lead to "bread lseek errors" on large filesystems. Thanks
35 to Bruce Lowekamp <lowekamp@users.sourceforge.net> for
36 reporting this bug and debugging the issue.
37
38 5. Made dump use the blkid library when searching for devices
39 by label or uuid instead of dump's own routines.
40
41 6. Corrected a bug in dump where a wrong LABEL=... line in
42 /etc/fstab could prevent dump from dumping unrelated
43 filesystems. Thanks to Bruce Lowekamp
44 <lowekamp@users.sourceforge.net> for reporting the bug.
45
46 Changes between versions 0.4b35 and 0.4b36 (released April 21, 2004)
47 ====================================================================
48
49 1. Fixed dump compilation with old gcc versions. Thanks to
50 Mike Castle <dalgoda@users.sourceforge.net> for the patch.
51
52 2. Fixed some warnings (howmany, roundup, powerof2 redefined)
53 when compiling against a recent glibc version.
54
55 3. Fixed a bug in restore preventing the read of a dump tape
56 written with Solaris 7 ufsdump. Thanks to Patrick Higgins
57 <phiggins@transzap.com> for reporting the bug and providing
58 the test case.
59
60 4. Changed dump to enable the creation of volumes of different
61 sizes in a single run (make -B accept a list of values).
62 Patch contributed by Florian Zumbiehl <florz@gmx.de>.
63
64 5. Use the glibc provided minor() and major() macros instead
65 of our own bitmask implementation. This should be safe for
66 when the major/minor namespace will migrate to 32 bits.
67 Thanks to Zhang Jun <zhangjun@nanjing-fnst.com> for reporting
68 the bug.
69
70 6. Made explicit in the dump man page that dump will not create
71 a remote file, it will only write to an already existing one.
72
73 7. Another try at making size estimates better again.
74
75 8. Put back the inconditional running of the end-of-tape
76 script which was changed in 0.434 to be run only when -M
77 or multiple -f were NOT specified. Some users rely on this
78 feature even when it is combined with -M/-f.
79
80 9. Fixed restore when restoring huge backups (where rstdir...
81 temporary files are over 2GB). Thanks to
82 Raphael Jaffey <rjaffey@artic.edu> for reporting this,
83 debugging the issue and making the patch.
84
85 10. Made restore understand FreeBSD UFS2 tapes. Thanks to
86 David <vrtin@users.sourceforge.net> for submitting the bug
87 and providing a test case.
88
89 11. Made dump work with an arbitrary integer level (no more 10
90 levels only limitation). Thanks to Matthew
91 <msvincen@midway.uchicago.edu> for the patch.
92
93 Changes between versions 0.4b34 and 0.4b35 (released December 21, 2003)
94 =======================================================================
95
96 1. Added a note in the dump man page saying that the default
97 blocksize can be 32 if -d is used with a high density argument.
98 Thanks to Antonios Christofides <A.Christofides@itia.ntua.gr>
99 for the patch.
100
101 2. Fixed configure to correctly understand CPPFLAGS, CFLAGS,
102 LDFLAGS environment variables. Thanks to Arcady Genkin
103 <antipode@users.sourceforge.net> for reporting the bug.
104
105 3. Made -e/-E options of dump accept an unlimited number of inodes
106 to be excluded, rather than a hardcoded maximum. Thanks to
107 Dietrich Rothe <d-rothe@users.sourceforge.net> for the patch.
108
109 4. Updated the autoconf system to 2.50. Forced the -D_BSD_SOURCE
110 and -D_USE_BSD_SIGNAL defines in configure in order to solve
111 64bit build problems because quad_t is redifined with a
112 different signature. Thanks to Mike Harris <mharris@redhat.com>
113 for reporting this bug.
114
115 5. Made restore build on Solaris, making possible to
116 restore Linux's "enhanced" tapes. Thanks to Uwe Gohlke
117 <uwe@ugsoft.de> for the patch.
118
119 6. Made an extension in the dump tape format capable of saving
120 MacOSX specific inode extensions. Uwe Gohlke <uwe@ugsoft.de>
121 wrote the extension and contributed the restore code back
122 into this codebase. The same extension mechanism will be
123 used in the future to save ACLs...
124
125 7. Made rmt work correctly with regard to QFA and local/remote
126 files and tapes. The remote access will however work only
127 when the dump provided rmt version is used. If you want to
128 use another rmt server, please do not use the QFA feature.
129 Thanks to Uwe Gohlke <uwe@ugsoft.de> for the patch.
130
131 Changes between versions 0.4b33 and 0.4b34 (released April 18, 2003)
132 ====================================================================
133
134 1. Fixed the INSTALL file to reflect the actual install paths.
135 Thanks to David Ronis <ronis@ronispc.chem.mcgill.ca> for
136 reporting the bug.
137
138 2. Fixed the configure script to only check for headers presence
139 instead of trying to compile them. This should fix issues
140 with old build environments. Thanks to Kari Lempiainen
141 <kari@funky.pp.fi> for reporting the bug.
142
143 3. Fixed restore to correctly ignore sockets when comparing
144 a dump (as socket cannot be properly restored anyway). Thanks
145 to Gunther Reiszig <gunther@mit.edu> for reporting the bug.
146
147 4. Fixed restore to correctly access the archive file (-A argument)
148 even when using a remote tape. Thanks to Fabrice Bellet
149 <fabrice@bellet.info> for reporting the bug.
150
151 5. Fixed (again) handling of long (largefile) seeks in rmt.
152 Thanks to Fabrice Bellet <fabrice@bellet.info> for reporting
153 the bug.
154
155 6. Fixed restore corner case when dealing with large block sizes
156 dump is able to write now (-b 1024). Thanks to Fabrice Bellet
157 <fabrice@bellet.info> for reporting the bug.
158
159 7. Fixed a bug preventing dump to access a filesystem having
160 a label exactly 16 bytes in length. Thanks to <minduni@ti-edu.ch>
161 for reporting the bug.
162
163 8. Made dump store dump dates using explicit timezones, fixing a
164 problem with incremental dumps if the timezone is modified
165 between the dumps. Thanks to Philippe Troin <phil@fifi.org> for
166 the bug report and the patch.
167
168 9. Fixed a bug encountered when dumping individual files (not full
169 filesystems or directories) and dangling symbolic links happen
170 to be in the list of files. For as far as dump is concerned,
171 dangling symbolic links are allowed, and are dumped as is.
172 Thanks to Jin-su Ahn <jsahn@ee.snu.ac.kr> for reporting the
173 bug and providing the fix.
174
175 10. Fixed open and creation modes and permissions for QFA and
176 table-of-contents files in dump and restore. Thanks to
177 Philippe Troin <phil@fifi.org> for the patch.
178
179 11. Fixed the archive file descriptor handling enabling it to be 0.
180 This can happen in some cases when shell redirections are used.
181 Thanks to Philippe Troin <phil@fifi.org> for the patch.
182
183 12. Delayed the opening of archive file until after suid had been
184 dropped (fixing a possible security issue if dump is suid).
185 Thanks to Philippe Troin <phil@fifi.org> for the patch.
186
187 13. Fixed the 'S' command handling in the rmt client part.
188 Thanks to Philippe Troin <phil@fifi.org> for the patch.
189
190 14. Modified the end-of-tape script handling to print out statistics
191 (and stop the timer) before launching the eot script. Also, the eot
192 script does not get run anymore when using -M (which makes sense) or
193 when multiple tapes are listed on the command line
194 (-f tape0,tape1,tapen) (which also makes sense).
195 Thanks to Philippe Troin <phil@fifi.org> for the patch.
196
197 15. Relicensed dump/restore under the 'revised' BSD license, as per
198 ftp://ftp.cs.berkeley.edu/ucb/4bsd/README.Impt.License.Change.
199
200 16. Added LZO compression to dump. This new compression method has
201 the advantage of being super fast, thus not killing tape streaming
202 on slow machines. Thanks to Helmut Jarausch
203 <jarausch@igpm.rwth-aachen.de> for the patch and to
204 Markus Oberhumer <markus@oberhumer.com> for giving special permission
205 to include his miniLZO project (GPL licensed) in dump/restore.
206
207 17. Some small buffer overruns fixes in rmt. Thanks to Antonomasia
208 <ant@notatla.demon.co.uk> for reporting the bugs.
209
210 18. Added a special rmt version which can do encryption when writing
211 to tape. Read examples/encrypted_rmt/README for details on
212 how to enable and configure it. Thanks to Ken Lalonde
213 <ken@globalremit.com> for the patch.
214
215 19. Made dump work with 2.5 kernel end of tape early warning semantics.
216 Thanks to Kai Makisara <Kai.Makisara@kolumbus.fi> for the patch.
217
218 20. Fixed a bug which caused dump -w|-W not to work anymore, because
219 the fs_freq and fs_passno fields in /etc/mtab are always set
220 to 0 0. Thanks to Trent Piepho <xyzzy@speakeasy.org> for
221 reporting the bug.
222
223 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
224 =======================================================================
225
226 1. Added a note in the restore man page clarifying the question
227 'set the permissions on the current directory ?' asked by
228 restore at the end of treatment in -i and -x modes.
229
230 2. Fixed the endianess issues when reading compressed tapes.
231 Thanks to Dark Force <daq4th@users.sourceforge.net> for
232 reporting this bug and providing test cases.
233
234 3. Fixed the "ACL won't be dumped" warning message (which showed
235 an extra, unrelated error message). Thanks to Dragan Krnic
236 <dkrnic@lycos.com> for reporting this bug.
237
238 4. Made dump look first into /etc/mtab, then into /etc/fstab
239 when searching for filesystem paths. Also fixed some problems
240 caused by binding mounts or multiple block device mounts.
241 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
242 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
243 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
244 and #160305, Sourceforge bugs #618699 and #679832).
245
246 5. Made dump's -I option accept the value '0' meaning all the
247 read errors will be ignored. This can be useful when running
248 dump from unattended sessions (like cron jobs). Thanks to
249 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
250
251 6. Fixed the output of dump to indicate 'blocks' instead of
252 'tape blocks' in the various messages (blocks are always
253 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
254 made it clearly print the current blocksize at the start of
255 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
256 the suggestions.
257
258 7. Made rmt understand long (largefiles) seeks.
259
260 8. Fixed build with very old versions of libext2fs, where
261 EXT2_FT_* constants were undefined.
262
263 9. Made dump accept the dumpdates path on the command line
264 (-D file option) instead of using only the hardcoded one.
265 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
266 suggestion.
267
268 10. Enabled rmt, LFS, readline, QFA options by default in
269 ./configure. Updated the configure process (new versions
270 of config.guess, config.sub etc).
271
272 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
273 =======================================================================
274
275 1. Changed dump to use fcntl(F_SETLK) style locking instead
276 of flock() when locking the dumpdates file. With the old
277 locking scheme, a local user having read rights on the
278 dumpdates file could be able to do a Denial of Service attack
279 on dump. In order to lock the dumpdates file with the new
280 scheme, the user would need to have write access on the file.
281 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
282 reporting the bug (originally a bugtraq post).
283
284 2. Fixed interactive 'ls' which caused spurious errors warnings
285 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
286 <jorgen_ostling@users.sourceforge.net> for reporting this
287 bug.
288
289 3. Fixed dump's estimate when dealing with sparse inodes.
290
291 4. Modified dump to allow setting a blocksize bigger than 32kB
292 (raised the limit to 1024kB), because newer hardware needs
293 this for proper operation. Thanks to Dirk Traenapp
294 <dtraenapp@users.sourceforge.net> for reporting this.
295
296 5. Fixed a bug causing Dump to stop and report an error if an
297 inode number in the exclude file was followed by some amount
298 of whitespace. Thanks to Jeffrey Sofferin
299 <sofferin@users.sourceforge.net> for reporting this bug.
300
301 6. Fixed a bug which caused restore, in some particular cases,
302 to ask some 'scary' questions and leave a bunch of RSTTMP
303 directories behind when restoring incremental tapes. Thanks
304 to Philippe Troin <phil@fifi.org> for reporting this bug and
305 providing the test cases.
306
307 7. Changed the wording when inodes are excluded from dump:
308 replaced 'Added inode 7 to exclude list' with
309 'Excluding inode 7 (resize inode) from dump', as suggested
310 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
311
312 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
313 ===================================================================
314
315 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
316 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
317 <eros@lamel.bo.cnr.it> for reporting the bug.
318
319 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
320 ===================================================================
321
322 1. Made dump print out the ext2 logical block number in case of a read
323 error, which can be given as an argument to debugfs' ncheck command,
324 in order to find out the affected inode. Added note about this
325 usage in dump's man page.
326
327 2. Fixed a problem in restore when reading tapes written on big
328 endian machines with very old versions of dump. The patch was
329 contributed by George Helffrich <george@geology.bristol.ac.uk>.
330
331 3. Fixed the tape length calculation when using large tapes
332 and compression. Thanks to Georg Lippold
333 <g_lippold@sourceforge.net> for reporting the bug.
334
335 4. Added a new set of examples in dump_on_cd_2 directory, based
336 on dump_on_cd examples but somewhat enhanced, supporting DVD
337 media, and localized in english and german. Thanks to
338 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
339
340 5. Made dump save 32 bit UID/GID. Internally, this was achieved
341 by switching from the old BSD inode format to the new BSD
342 inode format, which means that the tape format was changed.
343 However, since all restore versions out there should
344 transparently support both inode formats, the change should
345 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
346 reporting the bug.
347
348 6. Fixed a lot of warnings in the code shown when compiling
349 with 'gcc -W'. Thanks to Matthias Andree
350 <matthias.andree@stud.uni-dortmund.de> for reporting this.
351
352 7. Fixed a small markup bug in the dump man page. Thanks to
353 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
354 patch.
355
356 8. Rewrote entirely the man pages using the tmac.an macro
357 package (Linux man page format) instead of the original BSD
358 format. They should be now cleaner and easier to modify.
359
360 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
361 ==================================================================
362
363 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
364 non Linux clients were misinterpreted. The description of the problem
365 (incompatible numbering in Linux mtio opcodes) is documented at
366 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
367 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
368 providing an excellent, cross-platform replacement for rmt in his
369 star package.
370
371 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
372 which caused restore to fail when the inode maps were not entirely
373 contained on the first volume (when using really small volumes or
374 when dumping a huge number of inodes).
375
376 3. Fixed a problem in dump, where files in subdirectories of directories
377 that have the nodump flag set (or being excluded with -e/-E) were
378 sometimes incorrectly dumped (instead of being excluded). The original
379 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
380 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
381 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
382 help testing the patch.
383
384 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
385 which makes one able to pipe the output of dump, by the net, to
386 a remote CD-burner server.
387
388 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
389 and use GNU's symbolic syntax over rmt instead of numerical values
390 to assure multiple platform compatibility.
391
392 6. Documented the -d option in restore.
393
394 7. Added a -v (verbose) mode to dump. For now it just prints the number
395 of the inode being dumped, but this could evolve in future versions
396 to include interesting debugging output.
397
398 8. Added a -o flag to restore, which automatically restores the current
399 directory permissions in -i or -x mode, without asking the operator
400 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
401 and Peter Samuel <peters@e-smith.com>.
402
403 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
404 ====================================================================
405
406 1. Fixed a bug in the QFA feature which could have caused
407 a bad tape positionning on restore (causing extra delays in
408 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
409
410 2. Added a small note in the dump man page specifying that
411 there should be no white space between the option letter and
412 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
413
414 3. Made restore work with older versions of the readline library,
415 by checking in configure for several extended symbols. Restore
416 can now be compiled with a readline as old as the 2.0 release
417 (though it may be a good idea to upgrade it to a more recent
418 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
419 reporting the build failures.
420
421 4. Fixed a performance problem with the QFA file creation in
422 dump, which made unnecessary seeks on the tape slowing down
423 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
424 reporting this issue.
425
426 5. Removed the inclusion of some kernel headers in the dump
427 source, which prevented the compile in some kernel/glibc
428 headers/architecture combination. Thanks to Bdale Garbee
429 <bdale@gag.com> for reporting the bug.
430
431 6. Added the appropriate error message when dump fails to
432 open the output file for writing. Thanks to Amith Varghese
433 <amithv@yahoo.com> for reporting this bug.
434
435 7. Made restore able to understand large Solaris ufsdump tapes
436 (containing inodes bigger than 4194304). Sun have introduced
437 an "extension" to the dump tape format when dealing with
438 those inodes, which was uncorrectly handled by Linux restore.
439 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
440 providing a test case.
441
442 8. Added the -m parameter to dump which optimises the output for
443 inodes having been changed but not modified since the last dump
444 ('changed' and 'modified' have the meaning defined in stat(2)).
445 For those inodes, dump will save only the metadata, instead of
446 saving the entire inode contents. Inodes which are either
447 directories or have been modified since the last dump are saved
448 in a regular way. Uses of this flag must be consistent, meaning
449 that either every dump in an incremental dump set have the flag,
450 or no one has it.
451
452 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
453 =======================================================================
454
455 1. Fixed behaviour of dump when exceeding resource limits
456 (SIGXFSZ treatment).
457
458 2. Added the -L flag to restore to allow the user to specify a
459 maximal allowed number of miscompares when using restore
460 with the -C option to check the backup.
461
462 3. Detailed the manual entry for the -N option of restore.
463
464 4. Added the -a flag to restore to make able doing unattended
465 restores in -i or -x mode (automatically walks through the
466 multiple dump volumes).
467
468 5. Extended the QFA mode to work with local files and/or
469 remote tapes and files. This way, restore can know in advance
470 the tape number and the offset for the inodes to extract and
471 can minimize the extraction time by seeking directly to the
472 good tape/offset.
473
474 6. Added the -A <archive> option to both dump and restore,
475 which makes dump to archive a dump table-of-contents in
476 the specified file to be used by restore to determine
477 whether a file is in the dump file that is being restored.
478 (the archive file syntax is also compatible with the
479 Solaris ufsdump generated one).
480
481 7. Small fix in restore making it able to read some (broken ?)
482 Solaris ufsdump tapes.
483
484 8. Fixed dump to correctly recognise the root filesystem when using
485 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
486 for reporting this bug.
487
488 9. Added the -P <file> option to restore to create a
489 Quick File Access file from an already made dump. Patch
490 contributed by Uwe Gohlke <uwe@ugsoft.de>.
491
492 10. Made restore compile and run on Solaris, making it a
493 possible replacement for the standard ufsrestore. Port was
494 contributed by Uwe Gohlke <uwe@ugsoft.de>.
495
496 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
497 =====================================================================
498
499 1. Added a set of backup scripts from Eugenio Diaz
500 <getnito@yahoo.com> in the examples section. It features
501 automatic (cron based) full and incremental dumping of
502 several filesystems on a separate filesystem.
503
504 2. Fixed a off-by-one miscalculation which disabled dumping
505 a one letter subdirectory. Thanks to Chris Siebenmann
506 <cks@utcc.utoronto.ca> for reporting the bug.
507
508 3. Fixed several restore bugs occuring when trying to
509 restore some missing files on the tape. Thanks to Chris
510 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
511
512 4. Fixed --with-ldopts configure argument passing, installing from
513 a separate object directory, makefile cleanups contributed
514 by <splite@purdue.edu>.
515
516 5. Fix a bug which could caused, in some conditions, the highest
517 number inode of a filesystem, to not be dumped. Many thanks
518 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
519 chase this bug.
520
521 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
522 =======================================================================
523
524 1. Added a mini howto from Patrick Walsh in the examples
525 directory.
526
527 2. Minor man pages syntax corrections. Thanks to
528 Chris Danis <screechco@home.com> for reporting the bugs.
529
530 3. Added a script from David B. Peterson <dave@toppledwagon.com>
531 to the examples section. It features dumping several
532 filesystems to a remote tape drive upon ssh.
533
534 4. Added a patch provided by Richard Jones <rich@annexia.org>
535 which allows BRADEMAX (number of read errors tolerated by
536 dump) to be adjusted using the -I option.
537
538 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
539 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
540 patch.
541
542 6. Add the -l option to restore to specify if, when doing a
543 remote restore, the file used to access the backup is a
544 regular file (the defaults being a tape drive). Restore needs
545 to know this information when reading a remote compressed
546 dump. Previously, this information was autodetected, but
547 the autodetection code fails (with ioctl: Inappropriate ioctl
548 for device) when using a non Linux remote box. Thanks to
549 many users and especially to Eros Albertazzi
550 <eros@lamel.bo.cnr.it> for reporting this.
551
552 7. Found a workaroung for the dump deadlock problem (3 childs
553 stuck in pause(), father in read()). The workaround seems
554 to work for me and several beta-testers. If it doesn't work
555 for you, please report back.
556
557 8. Updated the RPM spec file (BuildPrereq, URL etc).
558
559 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
560 ========================================================================
561
562 1. Fixed the permissions of a newly created QFA file by dump.
563
564 2. Cleaned up the source of dump (the external variables
565 definition was a complete mess, making possible to have
566 objects overlap).
567
568 3. Fixed restore to use the full tape volume path when doing
569 a compare (since it changes the working directory to the
570 filesystem being compared in the process).
571
572 4. Added the -q option to dump which makes dump abort
573 whenever it needs operator attention. It should be
574 easier to use dump in scripts with this option.
575
576 5. Detect the use of incompatible options to dump and
577 refuse them (like -a and -B options together).
578
579 6. Added bzip2 compression to dump/restore (use option -j level
580 to select it). Note that you will need a restore version
581 >= 0.4b24 in order to restore a bzip2 compressed dump.
582 The same warning as for the zlib compression applies:
583 the tape format of a bzip2 dump is not compatible with the
584 original BSD tape format.
585
586 7. Fixed a overflow problem in dump corrupting the dump when
587 very large files were encountered. Thanks to Vladimir Nadvornik
588 for the bug report.
589
590 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
591 kernel buffer/page cache before starting the dump, helping
592 a bit those who use dump on mounted filesystems. Thanks to
593 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
594 for this suggestion.
595
596 9. Updated the RPM spec file following the RedHat changes
597 (dynamically linked binaries now in /usr/sbin etc).
598
599 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
600 which enables restore to recognise multi volume compressed dumps
601 done on CD. Included his scripts for dump (which pipe the dump
602 output directly into cdrecord) and restore. There is now
603 possible to to backups to CD on the fly!
604
605 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
606 ===================================================================
607
608 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
609 Marcus Meissner (Caldera International Security Dept.).
610
611 2. Implement the Sun rmt extended protocol. Patch submitted
612 by Ian Gordon <iangordon@users.sourceforge.net>.
613
614 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
615 the linux kernel header. This ensures that dump always has the
616 latest version of this file. Patch submitted by
617 Andreas Dilger <adilger@turbolinux.com>.
618
619 4. Report any filesystem present in either /etc/fstab with a
620 positive passno or /etc/dumpdates in dump -w output.
621 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
622
623 5. Fixed the looping problem in dump introduced in the
624 previous version.
625
626 6. Changed the -B option of dump to limit the size of
627 _compressed_ output per volume if compression is on.
628 Patch contributed by Helmut Jarausch
629 <jarausch@igpm.rwth-aachen.de>. Note however that, since
630 it is impossible to predict the size of the compressed
631 data before doing the compression, the -B limit is a bit
632 conservative.
633
634 7. Fixed a bug in reading the operator typed file/tape path for
635 the next volume in restore.
636
637 8. Implemented a "-F script" option to restore which permits the
638 user to specify a script which will be launched at the
639 beginning of each tape, useful for automatic programming of
640 tape changers for example. See the restore man page for the
641 script parameters and return codes.
642
643 9. Small fix for the QFA routines provided by Uwe Gohlke
644 <uwe@ugsoft.de>, and some recommendations for QFA uses in
645 the man pages.
646
647 10. Fixed the multivolume restoring where making a mistake
648 to the 'Mount next tape' prompt caused several blocks to
649 be lost.
650
651 11. Enhanced the -e option of dump to take as a parameter a
652 comma separated list of inode numbers.
653
654 12. Added the -E option to dump which specify a file containing
655 inode numbers to exclude from the dump.
656
657 13. Fixed the compressed multi-volume dump + restore.
658
659 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
660 ==================================================================
661
662 1. Made dump report the number of blocks dumped per volume.
663 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
664
665 2. Fix a bug in dump where waiting too much at the 'change volumes'
666 question caused the volume to be rewritten. Thanks to
667 Graham Reed <greed@users.sourceforge.net> for reporting the
668 bug and providing a patch.
669
670 3. Added a compression option to dump, contributed by Jerry
671 Peters <gapeters@worldnet.att.net>.
672
673 WARNING: the compression works only on files/pipes and
674 drives supporting variable block size.
675
676 WARNING2: tapes created using the compression feature are
677 incompatible with the standard dump tape format, and a
678 version of restore >= 0.4b22 will be needed for extraction.
679
680 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
681 architectures. Thanks to Paul Slootman <paul@debian.org> for
682 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
683 it upstream.
684
685 5. Many cleanups (CPP defines, const char warnings, check of
686 ext2fs COMPAT flags, time_t cleanups, added libext2 version
687 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
688
689 6. Made --prefix option work in configure. All the install path
690 are now based on the configure parameters.
691
692 7. Added the Quick File Access mode in dump/restore, contributed
693 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
694 a file tape position for each inode, and this file is used by
695 restore (if called with parameter Q and the filename)
696 to directly position the tape at the file restore is currently
697 working on. This saves hours when restoring single files from
698 large backups, saves the tapes and the drive's head. Use
699 --enable-qfa option of configure to compile in the QFA support.
700
701 8. Added the possibility to dump several files and directories
702 in a single invocation of dump. Thanks to Uwe Gohlke
703 <uwe@ugsoft.de> for implementing this option.
704
705 9. Fixed the dumping and restoring of files with holes
706 (files ending with holes didn't get dumped or restored
707 correctly).
708
709 10. Fixed a socket descriptor leak in dump, which leaved opened
710 3 file descriptors per dump process (and there is one dump
711 process per tape).
712
713 11. Fixed dump large file system support, by explicit use of
714 open64/lseek64/etc functions (explicit use needed because
715 e2fsprogs libraries don't behave well when compiled with
716 FILE_OFFSET_BITS=64).
717
718 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
719 ======================================================================
720
721 1. Fixed some bugs in the dump -w|-W logic introduced by
722 the previous version. Thanks to Andreas Dilger
723 <adilger@turbolinux.com> for his help on this one.
724
725 2. Fixed again a compilation problem when using old e2fs
726 headers (filesystem label related). Thanks to many users
727 who reported this stupid error.
728
729 3. Fixed a build problem on old lib5 based systems dealing with
730 _PATH_MNTTAB being undefined. Thanks to John Adams
731 <johna@onevista.com> for reporting the bug.
732
733 4. Improved the error detection in restore when creating
734 the temporary files in TMPDIR. Now restore will corectly
735 report a 'no space left on device' error instead of
736 strange errors which could imply an archive corruption.
737 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
738 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
739
740 5. Added the throughput information in dump's progression
741 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
742 for the patch.
743
744 6. Use libext2fs's inode scanning routines, which are particularly
745 robust when dealing with errors (especially when having some
746 bad blocks on the filesystem). Many thanks to Colin
747 <colin@horizon.com> for reporting the bug, investigating
748 the issues, finding a workaround, writing the patch and
749 fully testing it... (of course, if this patch will break
750 anything, he is to blame for :-)).
751
752 7. Made dump and restore LFS aware. Dump can dump a filesystem
753 containing large files, generate a large file on output and
754 restore can restore them. This feature is not enabled by
755 default, use --enable-largefile option of configure to enable
756 it (you need to have a LFS aware glibc though). Thanks to
757 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
758 and to Theodore T'so <tytso@valinux.com> for his always
759 useful thoughts.
760
761 8. Made dump ask upon a tape write error if it should rewrite
762 the current volume (assume this is a bad tape) or if it should
763 assume an end-of-tape condition (useful for tape drives which
764 behaves badly at the end of the tape). Thanks to Andreas
765 Dilger <adilger@turbolinux.com> for the suggestion.
766
767 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
768 =======================================================================
769
770 1. Fixed a small compilation problem due to a change
771 in the definintion of the struct sigaction in
772 glibc 2.0 / libc5. Thanks to Gunther Schlegel
773 <schlegel@riege.de> for reporting the bug and to
774 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
775 a fix.
776
777 2. Modified the label and uuid analysis in order to be
778 self-contained (independant of kernel/libc headers). This
779 should fix the compile with older kernel/libc headers and
780 will preserve the functionality. Thanks to Bernhard Erdmann
781 <bernhard.erdmann@gmx.de> for reporting the bug.
782
783 3. The 'exclude inode' option, if applied to a directory
784 inode, excludes now correctly all the leaves of this
785 directory. Thanks to John R. Dennison
786 <gerdesas@users.sourceforge.net> for reporting the bug.
787
788 4. Fixed the '-e' option to disable the possibility
789 to exclude the root inode (which causes the dumps to
790 be unreadable by restore). Prevented array overflow
791 when multiple -e options are used.
792
793 5. Fixed dump to correctly interpret a filesystem argument
794 which is a mountpoint and it is not an absolute path
795 (as specified in the fstab). Thanks to Bernhard R. Erdmann
796 <be@berdmann.de> for reporting the bug.
797
798 6. Made dump able to backup files larger than 2 GB. Note that
799 dump still doesn't cope with files larger than 4 GB.
800
801 7. Restore the real uid and gid before invoking an external
802 RSH program (big hole when dump or restore is suid root!).
803
804 8. Get the values for the default tape device and the location
805 of fstab file from the system headers. Thanks to
806 Andreas Dilger <adilger@turbolinux.com> for the patch.
807
808 9. Made dump -w|-W report all recognized filesystems
809 present in either /etc/fstab or /etc/dumpdates, and present
810 the list in the same order as in fstab file. Thanks
811 to Andreas Dilger <adilger@turbolinux.com> for the patch.
812
813 10. Made dump's -a (automatic end of tape detection) the
814 default. Specifying one of -B, -c, -d or -s options will
815 override the EOM detection. Thanks to Andreas Dilger
816 <adilger@turbolinux.com> for the patch.
817
818 11. Save the ext2 filesystem label into the dump volume label.
819 Specifying a volume label on the command line (-L option)
820 will override this feature. Thanks to Andreas Dilger
821 <adilger@turbolinux.com> for the patch.
822
823 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
824 =====================================================================
825
826 1. Fixed the signal handling in dump (which I broke in 0.4b17)
827 which was causing several strange dump failures (dump
828 hanged or segmentation faults etc).
829
830 2. Specified the default blocksize in the dump man page.
831
832 3. Changed two info messages of restore to be written on stdout
833 instead of stderr in order to leave stderr only for errors
834 or warnings. Thanks to Stephen Atwell
835 <satwell@urbana.css.mot.com> for the suggestion.
836
837 4. Corrected an off by one calculation which prevented
838 dumping a directory having a 1 character name length.
839 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
840 for reporting the bug.
841
842 5. Reinforce dump in order to allow only level 0 and no
843 -u option when dumping a subdirectory, and document
844 this in the man page. Thanks to Bernhard Erdmann
845 <bernhard.erdmann@gmx.de> for reporting the bug.
846
847 6. Fixed a small harmless bug in restore which caused
848 in some conditions hard links to be restored several
849 times (therefore generation some warning messages).
850 Thanks to Kenneth Porter <shiva@well.com> for
851 reporting the bug.
852
853 7. Updated the RPM spec file to the latest RedHat version,
854 providing FHS packaging and other cosmetic changes.
855 You will need RPM version >= 3.0.5 in order to build the RPM.
856
857 8. Updated the configure script to check for libtermcap
858 before checking for libreadline, because we need this
859 library in order to compile the readline support.
860
861 9. Made dump understand the LABEL= and UUID= notation
862 both in /etc/fstab and on the command line. Note that
863 you will need the /proc filesystem in order to use
864 these notations. Thanks to Erik Troan <ewt@redhat.com>
865 for providing the patch.
866
867 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
868 ===================================================================
869
870 1. Fixed a potential buffer overflow in restore. Thanks
871 to Stan Bubrouski <satan@fastdial.net> for reporting
872 the bug.
873
874 2. Fixed a readline-related bug which prevented
875 'cat DUMPFILE | restore -i -f -' from working. Thanks
876 to Charles Karney <karney@users.sourceforge.net>
877 for the bug report.
878
879 3. Changed a few "panic" into "exit", causing restore to
880 be more stable against some attacks (like the last one
881 reported on Bugtraq, although the last version of restore
882 was not vulnerable - just dumped core). Thanks to
883 Andreas Hasenack <andreas@conectiva.com.br> for reporting
884 the bugs to me.
885
886 4. Removed the suid-root bit on dump and restore in the
887 default build (and generated RPMs). It should be safer
888 now. Users who need the suid-root capabilities in order
889 to make network backups should read first the man page
890 of dump and enable the suid bit manually.
891
892 5. Added -ltermcap to the compile parameters for restore
893 when using readline support, in order to make the compile
894 process work on some systems (SuSE ?). Thanks to
895 Patrik Schindler <poc@pocnet.net> for reporting the bug.
896
897 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
898 ====================================================================
899
900 1. The -F script is called now *only* at the end of a tape,
901 not at the end of the dump. Thanks to Christian Haul
902 <haul@informatik.tu-darmstadt.de> for the bug report.
903
904 Normally, the device name and the current volume number
905 are passed on the command line of the script. However,
906 if you want the old style script (with no arguments
907 passed on the command line), you can enable it in
908 configure with the --enable-oldstylefscript.
909
910 2. Use posix signal handling to preserve dump functionality
911 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
912 the patch.
913
914 3. Made the exit code of restore in case of a 'C'ompare
915 command reflect the result of the compare. An exit status
916 of 0 means the dump archive is correct, 1 means tape errors,
917 2 means that some files were modified. Thanks to Kenneth Porter
918 <shiva@well.com> for the suggestion.
919
920 4. Made (finally) quotes work in interactive restore.
921
922 5. Small fixes in order to allow dump to compile with a
923 really old e2fsprogs version. Thanks to Ian Zimmerman
924 <itz@speakeasy.org> for the bug report.
925
926 6. Add GNU readline capabilities to interactive restore.
927 Use configure's --enable-readline flag to enable this feature.
928 Thanks to Patrik Schindler <poc@pocnet.net> for the
929 suggestion.
930
931 7. Do the compare on the fly in restore 'C' mode (this will
932 allow not to exhaust the available /tmp space when
933 ccmparing large files). Thanks to Kenneth Porter
934 <shiva@well.com> for the suggestion.
935
936 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
937 ====================================================================
938
939 1. Fixed some several duplicate 'DUMP: DUMP:' in the
940 output of dump.
941
942 2. Corrected the estimation of blocks to dump. Note that
943 this is still wrong for large directory inodes, due
944 to the size of a BSD directory entry on the tape
945 (estimation is based on the size of the ext2 directory,
946 which is smaller because it doesn't contain the null
947 character at the end).
948
949 3. Made dump report the total number of MB written to
950 tape at the end of run. Thanks to W. Reilly Cooley
951 <wcooley@nakedape.cc> for the patch.
952
953 4. Added the -X option to restore giving the possibility
954 to read the names of the files to be extracted/listed
955 from a text file (in addition of the command line).
956 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
957 patch.
958
959 5. Added the device name and the current volume number
960 as arguments to the end of tape script (dump -F option).
961
962 6. Made the multi-volume dump work again (after having
963 broken it in 0.4b15).
964
965 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
966 ===================================================================
967
968 1. Added a prompt command in interactive restore mode. Thanks
969 to Andreas Dilger <adilger@home.com> for the patch.
970
971 2. Fixed a buffer overflow problem in dump (caused by
972 not checking the size of the filesystem parameter).
973 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
974 reporting this on Bugtraq (and to several dump users
975 who forwarded me his mail).
976
977 3. Added the '-F script' option to dump in order to
978 launch a script at the end of each tape (to be used
979 with a tape changer, or to notify the sysadmin by
980 pager etc.).
981
982 4. Fixed a bug in restore compare code caused by the changes
983 I made in 0.4b14.
984
985 5. Fixed the treatment of options using the old BSD syntax
986 in both dump and restore.
987
988 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
989 =======================================================================
990
991 1. Fixed a bug in dump which may have caused invalid deleted
992 directories to be dumped out if they were deleted after the
993 mapping pass. This could occure on active filesystem and lead
994 to heap corruption (causing dump malfunction in many possible ways).
995 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
996 bug and submitting the patch.
997
998 2. Fixed the handling of the filesystem parameter in dump. A
999 '/mnt/disk' parameter caused the disk contents to be dumped,
1000 but a '/mnt/disk/' parameter caused the mountpoint directory
1001 to be dumped (generally an empty directory).
1002
1003 3. Improved the output of dump in order to tell which directory
1004 it is currently dumping (when dumping a subtree).
1005
1006 4. Added the '-e' exclude inode option to dump. Thanks to
1007 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
1008
1009 5. Added a REPORTING-BUGS file in order to provide a guide
1010 on how to correctly report a bug in dump/restore.
1011
1012 6. Fixed a restore problem when restoring a hard link to an inode
1013 having the immutable or append-only attribute set. Thanks to
1014 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
1015
1016 7. Fixed a compatibility problem between dumps made on little
1017 endian machines (the format was correct) and big endian
1018 machines (the format was incorrect). This fix break the
1019 compatibility with the older dumps made on big endian
1020 machines (sparc, m86k, ppc etc). For the first time in
1021 linux dump's history (I believe), the dumps made by *BSD,
1022 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
1023 so interchangeable. Thanks to Rob Cermak
1024 <cermak@ahab.rutgers.edu> for submitting the bug and
1025 helping me test the fix.
1026
1027 8. Fixed the way dump reports the remaining percent/time, if
1028 the number of blocks actually dumped exceeds the estimated
1029 number of blocks. Thanks to Jean-Paul van der Jagt
1030 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
1031
1032 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
1033 ======================================================================
1034
1035 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
1036 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
1037
1038 2. Removed the requirement to build the RPM as root from the
1039 spec file. Thanks to Christian Weisgerber
1040 <naddy@mips.rhein-neckar.de> for submitting this.
1041
1042 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
1043 causing dump to block when dumping really huge filesystems.
1044 Many thanks to Patrik Schindler <poc@pocnet.net> for
1045 helping me find this bug.
1046
1047 4. Fixed the treatment for an interrupt signal when dump access
1048 the remote tape through RSH. Thanks to Christian Weisgerber
1049 <naddy@mips.rhein-neckar.de> for providing the patch.
1050
1051 5. Fixed a bug which was causing dump/restore to display
1052 garbage characters instead of the remote host name.
1053
1054 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
1055 =====================================================================
1056
1057 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
1058 <kukuk@suse.de> for submitting the patch.
1059
1060 2. Fix for the exit code when using the size estimate option of
1061 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
1062 submitting the patch.
1063
1064 3. Handle EINTR in atomical reads/writes in dump, which was causing
1065 dump to fail on some systems. Thanks to Eric Jergensen
1066 <eric@dvns.com> for reporting the bug and submitting the patch.
1067
1068 4. Handle more than 16 characters for the device names in dumpdates.
1069 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
1070 tracking down the problem and proposing the solution.
1071
1072 5. Fixed a bug in dump which prevented the creation of the
1073 dumpdates file when doing a 0-level dump without already
1074 having a dumpdates file. Thanks to Patrik Schindler
1075 <poc@pocnet.net> for reporting the bug.
1076
1077 6. Changed the way dump 'S' flag reports the size estimate
1078 from number of blocks into bytes (making it compatible
1079 with the Solaris version, and simplifying things for
1080 amanda users). Thanks to Jason L Tibbitts III
1081 <tibbs@math.uh.edu> for reporting the bug.
1082
1083 7. Fixed a compatibility problem in linux/alpha dump tape format.
1084 Now the linux/alpha dump are (again) compatible with the
1085 other dump formats. But this breaks compatibility with
1086 older dumps made on alpha. Thanks to Mike Tibor
1087 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
1088
1089 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
1090 ======================================================================
1091
1092 1. Added a '--enable-kerberos' to configure.
1093
1094 2. Added a 'S' option to dump which determines the amount of space
1095 that is needed to perform the dump without actually doing it, similar
1096 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
1097 <cermak@ahab.rutgers.edu>.
1098
1099 3. Added a 'M' multi-volume option to dump and restore which enables
1100 dumping to multiple files (useful when dumping to an ext2
1101 partition to make several dump files in order to bypass the 2GB
1102 file size limitation). The 'f' argument is treated as a prefix and
1103 the output files will be named <prefix>001, <prefix>002 etc. With
1104 the 'M' flag, restore automatically selects the right file without
1105 asking to enter a new tape each time.
1106
1107 4. Fixed a memory leak which was causing dump to grow very big
1108 (270MB when dumping a 10GB filesystem...). Thanks to Jason
1109 Fearon <jasonf@netrider.org.au> for reporting the bug.
1110
1111 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
1112 ======================================================================
1113
1114 1. Make configure test if the system glob routines support
1115 extended syntax (ALTDIRFUNC). If not, use the internal glob
1116 routines instead of system ones. Thanks to Bernhard Sadlowski
1117 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
1118 and helping me resolve this and other minor libc5 compiling
1119 glitches.
1120
1121 2. Fix a problem when dumping a ext2fs with the 'filetype'
1122 feature enabled. Thanks to Patrick J. LoPresti
1123 <patl@cag.lcs.mit.edu> for reporting the bug and to
1124 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
1125
1126 3. Made the nodump flag work on directories. A directory which
1127 has the nodump flag gets never dumped, regardless of its
1128 contents.
1129
1130 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
1131 which allows dump on an active ext3 filesystem. However, this
1132 is a "quick and dirty" patch which enables backup of an ext3
1133 filesystem through the ext2 compatibility (by ignoring the
1134 NEEDS_RECOVERY bit). The journal file is not recognized and
1135 it is dumped (it should not).
1136
1137 5. Test the superblock compatibility flags when dumping, in order
1138 to be sure that we know how to deal with specific features.
1139
1140 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
1141 ====================================================================
1142
1143 1. Use lchown instead of chown, fixing a possible security problem
1144 when restoring symlinks (a malicious user could use this
1145 to deliberately corrupt the ownership of important system files).
1146 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
1147 this and providing the patch.
1148
1149 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
1150 ====================================================================
1151
1152 1. Put dump sources under CVS, added Id tags in all files so
1153 one can use 'ident' on binary files.
1154
1155 2. Added the dump/restore version in the usage text so one can
1156 easily verify the version he is using.
1157
1158 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
1159 a va_start/va_end problem on linux-ppc (always call va_start
1160 va_end in pairs each time we use a vararg function).
1161
1162 4. Added again the DT_* constants because old libc does not
1163 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
1164 for submitting the bug report.
1165
1166 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
1167 this should enable dumping big (huge) filesystems.
1168
1169 6. Added the RSH environment variable in order to be able to
1170 use a rsh replacement like ssh when doing remote backups (and
1171 bypass the security limitations of rcmd). Now you can do remote
1172 backups without being root (or making dump setuid root).
1173
1174 7. Modified again the way dumpdates works. For incremental dumps,
1175 we need to read dumpdates even if we are not using 'u' option.
1176 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
1177 this should work.
1178
1179 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
1180 ===================================================================
1181
1182 1. Removed the 'k' flag from the restore 'about' text if kerberos
1183 was not compiled in.
1184
1185 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
1186 to them (fsetflags takes a char*, setflags an open fd!).
1187
1188 3. (f)setflags is called only if the flags aren't empty. If the
1189 file is a special file, a warning is printed, because changing
1190 flags implies opening the device. Normally, a special file
1191 should not have any flag... (Debian bug #29775, patch provided
1192 by Abhijit Dasgupta <abhijit@ans.net>).
1193
1194 4. Made possible to dump a file system not mentioned in /etc/fstab.
1195 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
1196
1197 5. Changed the default behaviour to not create dumpdates
1198 unless 'u' option is specified. Removed the old "debian-patch"
1199 which provided the same thing. (Debian bug #38136, #33818).
1200
1201 6. Removed all those dump*announce, since they were getting old...
1202
1203 7. Added warning messages if dumpdates does not exist and
1204 when an inferior level dump does not exist (except for a level 0
1205 dump).
1206
1207 8. Debugged the glob calls in interactive mode: restore used a
1208 dirent struct which was different from the /usr/include/dirent.h
1209 one (this used to work, is it a glibc2 change?), so none of the
1210 compat glob (which used /usr/include/dirent.h) or the system glob
1211 worked. Restore use now the system dirent (and the system
1212 DT_* constants), which are compatible with BSD ones.
1213
1214 9. Added a configure flag (--with-dumpdatespath) to specify
1215 the location of dumpdates file. By default, it is
1216 /etc/dumpdates.
1217
1218 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1219 included the current date/version in man pages.
1220
1221 11. Corrected the estimation of remaining time when
1222 the operator doesn't change the tapes quickly enough. This
1223 was an old bug, I thought I corrected it, and discovered
1224 that in fact it was corrected in two different places, so
1225 the results canceled each other...
1226
1227 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1228 ===================================================================
1229
1230 1. Integrated multiple patches from RedHat, Debian and SuSE:
1231
1232 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1233 isn't specified.
1234 - several fixes in the man pages.
1235 - update the default tape device to /dev/st0.
1236 - many updates for Linux Alpha (byte ordering, size_t etc).
1237 - buffer overruns.
1238 - use environment variable for TMPDIR (instead of /tmp).
1239 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1240 - workaround egcs bug (RedHat bugs #4281 and #2989).
1241 - wire $(OPT) throughout Makefile's.
1242
1243 2. Upgrade the dump revision to 1, making possible to dump filesystems
1244 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1245
1246 3. Fix some compile warnings, prototype all functions.
1247
1248 4. Use glibc err/glob instead of internal compatibility
1249 routines (only if available).
1250
1251 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1252 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1253
1254 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1255 ======================================================================
1256
1257 1. Integrated the changes from FreeBSD-3.1-RELEASE
1258 (mostly bug fixes, buffer overruns, dump has now an "automatic
1259 tape length calculation" flag, dump/restore can use kerberos now
1260 (this is NOT tested), use environment variables for TAPE and
1261 RMT etc.).
1262
1263 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1264
1265 3. Corrected a bug in restore when using 'C' option with multi-volumes
1266 tapes (files splited accros two tapes give "size changed" errors
1267 when comparing).
1268
1269 4. Corrected the long standing bug when dumping multiple tapes.
1270 This works for me, needs further testing.
1271
1272 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1273 ====================================================================
1274
1275 1. Dump now runs correctly on kernels 2.1.x
1276 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1277
1278 Changes between versions 0.4b2 and 0.4b3
1279 ========================================
1280
1281 1. Use realpath() if available
1282
1283 2. Report statistics
1284
1285 Changes between versions 0.4b1 and 0.4b2
1286 ========================================
1287
1288 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1289 the patch)
1290
1291 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1292
1293 3. Fixed configure.in to correctly handle --enable-rmt
1294
1295 Changes between versions 0.3 and 0.4b1
1296 ======================================
1297
1298 1. Integrated the changes from 4.4BSD-Lite2
1299
1300 2. Integrated the patches from Debian and Red Hat
1301
1302 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1303
1304 4. Changed dump to use the Ext2fs library to get block addresses. This
1305 should solve the endianness problem on SparcLinux.
1306
1307 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1308 distribution's one) to use autoconf
1309
1310 6. Fixed a few minor bugs
1311
1312 Changes between versions 0.2e and 0.2f
1313 ======================================
1314
1315 1. Added the creation of named pipes (fifos) in restore.
1316
1317 2. Added the -N flag in the restore manual page.
1318
1319 3. Added the file kernel.patch which contains the llseek() optimization
1320 patch for 1.2.x kernels.
1321
1322 4. Fixed a bug in the restoration of symbolic links: owner and group were
1323 not restored.
1324
1325 5. Integrated some changes from FreeBSD 2.2.
1326
1327 6. Added a call to ftruncate() after restoring each file to restore
1328 correctly files ending by a hole.
1329
1330 Changes between versions 0.2d and 0.2e
1331 ======================================
1332
1333 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1334 with owner = root (0) and group = root/wheel/whatever (0).
1335
1336 Changes between versions 0.2c and 0.2d
1337 ======================================
1338
1339 1. Dump is now able to backup 2GB+ filesystems.
1340
1341 2. Dump and restore can now be linked as static binaries.
1342
1343 Changes between versions 0.2b and 0.2c
1344 ======================================
1345
1346 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1347
1348 Changes between versions 0.2a and 0.2b
1349 ======================================
1350
1351 1. Really fixed the bug that I should have corrected in 0.2a.
1352
1353 2. Enabled optimization again.
1354
1355 Changes between versions 0.2 and 0.2a
1356 =====================================
1357
1358 1. Disabled the optimization during compilation.
1359
1360 Changes between versions 0.1 and 0.2
1361 ====================================
1362
1363 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1364 the fs_type field (actually, I modified the file fstab.c to make it
1365 use the mntent functions).
1366
1367 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1368 made by dump should be readable by the BSD restore and backups made
1369 by the BSD dump should be readable by restore. Unfortunately, this
1370 means that the dump format has changed between version 0.1 and version
1371 0.2 :-(
1372
1373 3. Dump is now able to backup a subtree, it is no longer limited to whole
1374 filesystems like the BSD version.
1375
1376 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1377 than 2 GB.
1378
1379 Changes between versions 0.0 and 0.1
1380 ====================================
1381
1382 1. Now create links rdump and rrestore during the `make install' step.
1383
1384 2. Linux port specific bugs added to the manual pages
1385
1386 3. Incorrect estimation of the number of tapes blocks fixed when doing
1387 incremental backups.
1388
1389 4. Better ls-like format in restore in interactive mode.