Force -a when in 'restore -C' mode.
[dump.git] / CHANGES
1 $Id: CHANGES,v 1.290 2008/04/17 15:26:46 stelian Exp $
2
3 Changes between versions 0.4b41 and 0.4b42 (released xxxxxxxxxxxxxxx)
4 =====================================================================
5
6 1. Fix printout of the 'Connection to' message in dump/restore.
7 Thanks to Dale Wiles <dwiles@users.sourceforge.net> for the
8 bug report and its fix.
9
10 2. Fix dump -w/-W output for filesystems having the last backup
11 level equal to 0. Thanks to Pascal Bouchareine
12 <kalou@users.sourceforge.net> for reporting the bug.
13
14 3. Enable restore to handle restoring onto a different SELinux
15 policy from the dump. Thanks to Tony Nelson
16 <tony_nelson@users.sourceforge.net> for the patch.
17 (see http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=189845
18 for details)
19
20 4. Fix EA set failures when restoring immutable files. Thanks
21 to Andrew Kroeger for the patch.
22
23 5. Fix "mode file too large" errors when restoring huge backups.
24 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
25 report.
26
27 6. Add the ability to link against libtinfo library if present,
28 instead of libtermcap or libncurses libraries. Thanks to
29 Adam Tkac <atkac@redhat.com> for the patch.
30
31 7. When comparing a backup, do not consider to be an error the fact
32 that we encounter files created while dump was in progress.
33 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
34 report.
35
36 8. Force '-a' to be the default value when running 'restore -C'.
37 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
38 report.
39
40 Changes between versions 0.4b40 and 0.4b41 (released January 2, 2006)
41 =====================================================================
42
43 1. Fix restore of dumped Access Control Lists. The previous
44 code for EA works fine, but ACLs needed conversion from the
45 ext2/3 disk format to posix_acl format before restoring.
46
47 2. Fix some issues with restoration of EA on big endian
48 platforms.
49
50 3. Fix restore when the symtab is over 2GB in size.
51
52 4. Made the directory hash indexing an optional feature,
53 accessible by the '-H' option of restore, and disable it
54 by default.
55
56 5. Fixed dump to not include extended attributes information
57 in the toc (archive) file which confused restore -t.
58
59 Changes between versions 0.4b39 and 0.4b40 (released May 2, 2005)
60 =================================================================
61
62 1. Changed restore to emit warnings (instead of emitting a fatal
63 error) if a file (or a directory) is unavailable for a
64 comparision (if the user doesn't have the necessary permissions
65 to access it for example). Thanks to Kenneth Porter
66 <shiva@sewingwitch.com> for the bug report.
67
68 2. Re-done the 'do not save directory entries to non-dumped inodes
69 (excluded from dump)' feature. The previous implementation
70 worked well for excluded directories but not for regular files.
71 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
72 report.
73
74 3. Fixed a bug in dump where the tape size was miscalculated when
75 the user used -d/-s to specify the tape characteristics. Thanks
76 to Philip Goisman <goisman@physics.arizona.edu> for reporting
77 the bug.
78
79 4. Fixed another bug introduced in restore with the hashtree
80 implementation. This one caused restore to stop saying
81 "removenode: non-empty directory" in some cases.
82
83 5. Added support for dumping and restoring ext2/3 extended
84 attributes (EA), like the access control lists (ACL) or
85 the security labels used by SELinux.
86
87 Changes between versions 0.4b38 and 0.4b39 (released January 21, 2005)
88 ======================================================================
89
90 1. The newly added dump_on_cd_3 example was buggy, replace it
91 with an updated version from Andrew Basterfield
92 <bob@cemetery.homeunix.org>.
93
94 2. Made restore to chdir() back into the initial directory when
95 dumping core while aborting a comparision operation. The
96 previous behaviour was to write the corefile at the root of
97 the directory being compared, which could very well be
98 read only and preventing the corefile generation. Thanks
99 to Kenneth Porter <shiva@sewingwitch.com> for the bug report.
100
101 3. Silenced the failure to call fgetflags() when comparing an
102 entry which has no ext2 attributes (as in lsattr()).
103
104 4. Fix a brown paper bug in restore -C which broke restore and
105 caused modifications on the filesystem being compared
106 (directories containing a file with the same name as the
107 directory get renamed to RSTTMP...). Thanks to Kenneth Porter
108 <shiva@sewingwitch.com> for finding the bug and helping me
109 reproduce it.
110
111 5. Made restore -C force the -N flag (no writing allowed on
112 the disk) in order to prevent more bugs like the above one.
113
114 Changes between versions 0.4b37 and 0.4b38 (released January 7, 2005)
115 =====================================================================
116
117 1. Fix a couple of troff syntax bugs in the man pages.
118 Thanks to Eric Raymond <esr@thyrsus.com> for the patch.
119
120 2. Made restore use either libncurses or libtermcap, depending
121 on which one is available at configure time.
122
123 3. Fixed restore negative size display bug when comparing a
124 dump containing files over 2GB. Thanks to Steve Bonds
125 <sbonds@users.sourceforge.net> for the bug report.
126
127 4. Do not save directory entries to non-dumped inodes
128 (excluded from dump). This will eliminate the 'missing
129 file' warnings when doing 'restore -C'.
130
131 5. Fix dump crash when backuping a huge (2TB) filesystem,
132 due to a bogus calculation on the inode map size.
133 Thanks to Kevin B. Haines <K.B.Haines@rl.ac.uk> for
134 submitting the bug and testing the fix.
135
136 6. Fix a problem in restore where the final \0 in the symbolic
137 link names could have been lost, generating corrupt filenames.
138 Thanks to Kyle Wilson <kyle.wilson@amd.com> for reporting the
139 bug.
140
141 7. Implemented a hash list for the directory names in restore.
142 The linear list used before caused problems in interactive
143 restores when dealing with directories having thousands of
144 entries. Thanks to Brian Ristuccia <bristuccia@starentnetworks.com>
145 for reporting the bug.
146
147 8. Improved restore -C, this time including the directory
148 attributes into the comparision.
149
150 9. Made restore understand tapes containing EA/ACLs (which will
151 be dumped by the next version of dump). In this version
152 extended attributes on the tape are ignored, for full EA/ACL
153 support wait for the next version or try the experimental EA
154 patch.
155
156 Changes between versions 0.4b36 and 0.4b37 (released July 7, 2004)
157 ==================================================================
158
159 1. Added the --enable-staticz configure option which enables
160 dump and restore to be linked with static versions of
161 libz and libbz2 (and dynamic versions of all other libraries).
162 This will make Debian users happy, because libz and libbz2
163 were the only needed libraries living in /usr, all the
164 others live in /lib. In case of system emergency, it is
165 better not to have to rely on an extra filesystem.
166 Thanks to Bdale Garbee <bdale@gag.com> for the suggestion.
167
168 2. Fix compilation on (at least the Linux Debian port to) AMD64.
169 (<ext2fs/ext2_types.h> defines some types (__s64 and __u64)
170 that are also defined by <linux/types.h> (<asm/types.h>) and
171 they conflict).
172
173 3. Make dump's reading of the dumpdates file a bit more robust,
174 preventing dump from crashing when the dumpdates file has
175 been modified by hand.
176
177 4. Fixed some offset calculations in dump code which could
178 lead to "bread lseek errors" on large filesystems. Thanks
179 to Bruce Lowekamp <lowekamp@users.sourceforge.net> for
180 reporting this bug and debugging the issue.
181
182 5. Made dump use the blkid library when searching for devices
183 by label or uuid instead of dump's own routines.
184
185 6. Corrected a bug in dump where a wrong LABEL=... line in
186 /etc/fstab could prevent dump from dumping unrelated
187 filesystems. Thanks to Bruce Lowekamp
188 <lowekamp@users.sourceforge.net> for reporting the bug.
189
190 Changes between versions 0.4b35 and 0.4b36 (released April 21, 2004)
191 ====================================================================
192
193 1. Fixed dump compilation with old gcc versions. Thanks to
194 Mike Castle <dalgoda@users.sourceforge.net> for the patch.
195
196 2. Fixed some warnings (howmany, roundup, powerof2 redefined)
197 when compiling against a recent glibc version.
198
199 3. Fixed a bug in restore preventing the read of a dump tape
200 written with Solaris 7 ufsdump. Thanks to Patrick Higgins
201 <phiggins@transzap.com> for reporting the bug and providing
202 the test case.
203
204 4. Changed dump to enable the creation of volumes of different
205 sizes in a single run (make -B accept a list of values).
206 Patch contributed by Florian Zumbiehl <florz@gmx.de>.
207
208 5. Use the glibc provided minor() and major() macros instead
209 of our own bitmask implementation. This should be safe for
210 when the major/minor namespace will migrate to 32 bits.
211 Thanks to Zhang Jun <zhangjun@nanjing-fnst.com> for reporting
212 the bug.
213
214 6. Made explicit in the dump man page that dump will not create
215 a remote file, it will only write to an already existing one.
216
217 7. Another try at making size estimates better again.
218
219 8. Put back the inconditional running of the end-of-tape
220 script which was changed in 0.434 to be run only when -M
221 or multiple -f were NOT specified. Some users rely on this
222 feature even when it is combined with -M/-f.
223
224 9. Fixed restore when restoring huge backups (where rstdir...
225 temporary files are over 2GB). Thanks to
226 Raphael Jaffey <rjaffey@artic.edu> for reporting this,
227 debugging the issue and making the patch.
228
229 10. Made restore understand FreeBSD UFS2 tapes. Thanks to
230 David <vrtin@users.sourceforge.net> for submitting the bug
231 and providing a test case.
232
233 11. Made dump work with an arbitrary integer level (no more 10
234 levels only limitation). Thanks to Matthew
235 <msvincen@midway.uchicago.edu> for the patch.
236
237 Changes between versions 0.4b34 and 0.4b35 (released December 21, 2003)
238 =======================================================================
239
240 1. Added a note in the dump man page saying that the default
241 blocksize can be 32 if -d is used with a high density argument.
242 Thanks to Antonios Christofides <A.Christofides@itia.ntua.gr>
243 for the patch.
244
245 2. Fixed configure to correctly understand CPPFLAGS, CFLAGS,
246 LDFLAGS environment variables. Thanks to Arcady Genkin
247 <antipode@users.sourceforge.net> for reporting the bug.
248
249 3. Made -e/-E options of dump accept an unlimited number of inodes
250 to be excluded, rather than a hardcoded maximum. Thanks to
251 Dietrich Rothe <d-rothe@users.sourceforge.net> for the patch.
252
253 4. Updated the autoconf system to 2.50. Forced the -D_BSD_SOURCE
254 and -D_USE_BSD_SIGNAL defines in configure in order to solve
255 64bit build problems because quad_t is redifined with a
256 different signature. Thanks to Mike Harris <mharris@redhat.com>
257 for reporting this bug.
258
259 5. Made restore build on Solaris, making possible to
260 restore Linux's "enhanced" tapes. Thanks to Uwe Gohlke
261 <uwe@ugsoft.de> for the patch.
262
263 6. Made an extension in the dump tape format capable of saving
264 MacOSX specific inode extensions. Uwe Gohlke <uwe@ugsoft.de>
265 wrote the extension and contributed the restore code back
266 into this codebase. The same extension mechanism will be
267 used in the future to save ACLs...
268
269 7. Made rmt work correctly with regard to QFA and local/remote
270 files and tapes. The remote access will however work only
271 when the dump provided rmt version is used. If you want to
272 use another rmt server, please do not use the QFA feature.
273 Thanks to Uwe Gohlke <uwe@ugsoft.de> for the patch.
274
275 Changes between versions 0.4b33 and 0.4b34 (released April 18, 2003)
276 ====================================================================
277
278 1. Fixed the INSTALL file to reflect the actual install paths.
279 Thanks to David Ronis <ronis@ronispc.chem.mcgill.ca> for
280 reporting the bug.
281
282 2. Fixed the configure script to only check for headers presence
283 instead of trying to compile them. This should fix issues
284 with old build environments. Thanks to Kari Lempiainen
285 <kari@funky.pp.fi> for reporting the bug.
286
287 3. Fixed restore to correctly ignore sockets when comparing
288 a dump (as socket cannot be properly restored anyway). Thanks
289 to Gunther Reiszig <gunther@mit.edu> for reporting the bug.
290
291 4. Fixed restore to correctly access the archive file (-A argument)
292 even when using a remote tape. Thanks to Fabrice Bellet
293 <fabrice@bellet.info> for reporting the bug.
294
295 5. Fixed (again) handling of long (largefile) seeks in rmt.
296 Thanks to Fabrice Bellet <fabrice@bellet.info> for reporting
297 the bug.
298
299 6. Fixed restore corner case when dealing with large block sizes
300 dump is able to write now (-b 1024). Thanks to Fabrice Bellet
301 <fabrice@bellet.info> for reporting the bug.
302
303 7. Fixed a bug preventing dump to access a filesystem having
304 a label exactly 16 bytes in length. Thanks to <minduni@ti-edu.ch>
305 for reporting the bug.
306
307 8. Made dump store dump dates using explicit timezones, fixing a
308 problem with incremental dumps if the timezone is modified
309 between the dumps. Thanks to Philippe Troin <phil@fifi.org> for
310 the bug report and the patch.
311
312 9. Fixed a bug encountered when dumping individual files (not full
313 filesystems or directories) and dangling symbolic links happen
314 to be in the list of files. For as far as dump is concerned,
315 dangling symbolic links are allowed, and are dumped as is.
316 Thanks to Jin-su Ahn <jsahn@ee.snu.ac.kr> for reporting the
317 bug and providing the fix.
318
319 10. Fixed open and creation modes and permissions for QFA and
320 table-of-contents files in dump and restore. Thanks to
321 Philippe Troin <phil@fifi.org> for the patch.
322
323 11. Fixed the archive file descriptor handling enabling it to be 0.
324 This can happen in some cases when shell redirections are used.
325 Thanks to Philippe Troin <phil@fifi.org> for the patch.
326
327 12. Delayed the opening of archive file until after suid had been
328 dropped (fixing a possible security issue if dump is suid).
329 Thanks to Philippe Troin <phil@fifi.org> for the patch.
330
331 13. Fixed the 'S' command handling in the rmt client part.
332 Thanks to Philippe Troin <phil@fifi.org> for the patch.
333
334 14. Modified the end-of-tape script handling to print out statistics
335 (and stop the timer) before launching the eot script. Also, the eot
336 script does not get run anymore when using -M (which makes sense) or
337 when multiple tapes are listed on the command line
338 (-f tape0,tape1,tapen) (which also makes sense).
339 Thanks to Philippe Troin <phil@fifi.org> for the patch.
340
341 15. Relicensed dump/restore under the 'revised' BSD license, as per
342 ftp://ftp.cs.berkeley.edu/ucb/4bsd/README.Impt.License.Change.
343
344 16. Added LZO compression to dump. This new compression method has
345 the advantage of being super fast, thus not killing tape streaming
346 on slow machines. Thanks to Helmut Jarausch
347 <jarausch@igpm.rwth-aachen.de> for the patch and to
348 Markus Oberhumer <markus@oberhumer.com> for giving special permission
349 to include his miniLZO project (GPL licensed) in dump/restore.
350
351 17. Some small buffer overruns fixes in rmt. Thanks to Antonomasia
352 <ant@notatla.demon.co.uk> for reporting the bugs.
353
354 18. Added a special rmt version which can do encryption when writing
355 to tape. Read examples/encrypted_rmt/README for details on
356 how to enable and configure it. Thanks to Ken Lalonde
357 <ken@globalremit.com> for the patch.
358
359 19. Made dump work with 2.5 kernel end of tape early warning semantics.
360 Thanks to Kai Makisara <Kai.Makisara@kolumbus.fi> for the patch.
361
362 20. Fixed a bug which caused dump -w|-W not to work anymore, because
363 the fs_freq and fs_passno fields in /etc/mtab are always set
364 to 0 0. Thanks to Trent Piepho <xyzzy@speakeasy.org> for
365 reporting the bug.
366
367 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
368 =======================================================================
369
370 1. Added a note in the restore man page clarifying the question
371 'set the permissions on the current directory ?' asked by
372 restore at the end of treatment in -i and -x modes.
373
374 2. Fixed the endianess issues when reading compressed tapes.
375 Thanks to Dark Force <daq4th@users.sourceforge.net> for
376 reporting this bug and providing test cases.
377
378 3. Fixed the "ACL won't be dumped" warning message (which showed
379 an extra, unrelated error message). Thanks to Dragan Krnic
380 <dkrnic@lycos.com> for reporting this bug.
381
382 4. Made dump look first into /etc/mtab, then into /etc/fstab
383 when searching for filesystem paths. Also fixed some problems
384 caused by binding mounts or multiple block device mounts.
385 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
386 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
387 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
388 and #160305, Sourceforge bugs #618699 and #679832).
389
390 5. Made dump's -I option accept the value '0' meaning all the
391 read errors will be ignored. This can be useful when running
392 dump from unattended sessions (like cron jobs). Thanks to
393 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
394
395 6. Fixed the output of dump to indicate 'blocks' instead of
396 'tape blocks' in the various messages (blocks are always
397 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
398 made it clearly print the current blocksize at the start of
399 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
400 the suggestions.
401
402 7. Made rmt understand long (largefiles) seeks.
403
404 8. Fixed build with very old versions of libext2fs, where
405 EXT2_FT_* constants were undefined.
406
407 9. Made dump accept the dumpdates path on the command line
408 (-D file option) instead of using only the hardcoded one.
409 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
410 suggestion.
411
412 10. Enabled rmt, LFS, readline, QFA options by default in
413 ./configure. Updated the configure process (new versions
414 of config.guess, config.sub etc).
415
416 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
417 =======================================================================
418
419 1. Changed dump to use fcntl(F_SETLK) style locking instead
420 of flock() when locking the dumpdates file. With the old
421 locking scheme, a local user having read rights on the
422 dumpdates file could be able to do a Denial of Service attack
423 on dump. In order to lock the dumpdates file with the new
424 scheme, the user would need to have write access on the file.
425 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
426 reporting the bug (originally a bugtraq post).
427
428 2. Fixed interactive 'ls' which caused spurious errors warnings
429 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
430 <jorgen_ostling@users.sourceforge.net> for reporting this
431 bug.
432
433 3. Fixed dump's estimate when dealing with sparse inodes.
434
435 4. Modified dump to allow setting a blocksize bigger than 32kB
436 (raised the limit to 1024kB), because newer hardware needs
437 this for proper operation. Thanks to Dirk Traenapp
438 <dtraenapp@users.sourceforge.net> for reporting this.
439
440 5. Fixed a bug causing Dump to stop and report an error if an
441 inode number in the exclude file was followed by some amount
442 of whitespace. Thanks to Jeffrey Sofferin
443 <sofferin@users.sourceforge.net> for reporting this bug.
444
445 6. Fixed a bug which caused restore, in some particular cases,
446 to ask some 'scary' questions and leave a bunch of RSTTMP
447 directories behind when restoring incremental tapes. Thanks
448 to Philippe Troin <phil@fifi.org> for reporting this bug and
449 providing the test cases.
450
451 7. Changed the wording when inodes are excluded from dump:
452 replaced 'Added inode 7 to exclude list' with
453 'Excluding inode 7 (resize inode) from dump', as suggested
454 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
455
456 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
457 ===================================================================
458
459 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
460 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
461 <eros@lamel.bo.cnr.it> for reporting the bug.
462
463 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
464 ===================================================================
465
466 1. Made dump print out the ext2 logical block number in case of a read
467 error, which can be given as an argument to debugfs' ncheck command,
468 in order to find out the affected inode. Added note about this
469 usage in dump's man page.
470
471 2. Fixed a problem in restore when reading tapes written on big
472 endian machines with very old versions of dump. The patch was
473 contributed by George Helffrich <george@geology.bristol.ac.uk>.
474
475 3. Fixed the tape length calculation when using large tapes
476 and compression. Thanks to Georg Lippold
477 <g_lippold@sourceforge.net> for reporting the bug.
478
479 4. Added a new set of examples in dump_on_cd_2 directory, based
480 on dump_on_cd examples but somewhat enhanced, supporting DVD
481 media, and localized in english and german. Thanks to
482 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
483
484 5. Made dump save 32 bit UID/GID. Internally, this was achieved
485 by switching from the old BSD inode format to the new BSD
486 inode format, which means that the tape format was changed.
487 However, since all restore versions out there should
488 transparently support both inode formats, the change should
489 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
490 reporting the bug.
491
492 6. Fixed a lot of warnings in the code shown when compiling
493 with 'gcc -W'. Thanks to Matthias Andree
494 <matthias.andree@stud.uni-dortmund.de> for reporting this.
495
496 7. Fixed a small markup bug in the dump man page. Thanks to
497 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
498 patch.
499
500 8. Rewrote entirely the man pages using the tmac.an macro
501 package (Linux man page format) instead of the original BSD
502 format. They should be now cleaner and easier to modify.
503
504 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
505 ==================================================================
506
507 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
508 non Linux clients were misinterpreted. The description of the problem
509 (incompatible numbering in Linux mtio opcodes) is documented at
510 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
511 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
512 providing an excellent, cross-platform replacement for rmt in his
513 star package.
514
515 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
516 which caused restore to fail when the inode maps were not entirely
517 contained on the first volume (when using really small volumes or
518 when dumping a huge number of inodes).
519
520 3. Fixed a problem in dump, where files in subdirectories of directories
521 that have the nodump flag set (or being excluded with -e/-E) were
522 sometimes incorrectly dumped (instead of being excluded). The original
523 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
524 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
525 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
526 help testing the patch.
527
528 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
529 which makes one able to pipe the output of dump, by the net, to
530 a remote CD-burner server.
531
532 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
533 and use GNU's symbolic syntax over rmt instead of numerical values
534 to assure multiple platform compatibility.
535
536 6. Documented the -d option in restore.
537
538 7. Added a -v (verbose) mode to dump. For now it just prints the number
539 of the inode being dumped, but this could evolve in future versions
540 to include interesting debugging output.
541
542 8. Added a -o flag to restore, which automatically restores the current
543 directory permissions in -i or -x mode, without asking the operator
544 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
545 and Peter Samuel <peters@e-smith.com>.
546
547 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
548 ====================================================================
549
550 1. Fixed a bug in the QFA feature which could have caused
551 a bad tape positionning on restore (causing extra delays in
552 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
553
554 2. Added a small note in the dump man page specifying that
555 there should be no white space between the option letter and
556 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
557
558 3. Made restore work with older versions of the readline library,
559 by checking in configure for several extended symbols. Restore
560 can now be compiled with a readline as old as the 2.0 release
561 (though it may be a good idea to upgrade it to a more recent
562 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
563 reporting the build failures.
564
565 4. Fixed a performance problem with the QFA file creation in
566 dump, which made unnecessary seeks on the tape slowing down
567 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
568 reporting this issue.
569
570 5. Removed the inclusion of some kernel headers in the dump
571 source, which prevented the compile in some kernel/glibc
572 headers/architecture combination. Thanks to Bdale Garbee
573 <bdale@gag.com> for reporting the bug.
574
575 6. Added the appropriate error message when dump fails to
576 open the output file for writing. Thanks to Amith Varghese
577 <amithv@yahoo.com> for reporting this bug.
578
579 7. Made restore able to understand large Solaris ufsdump tapes
580 (containing inodes bigger than 4194304). Sun have introduced
581 an "extension" to the dump tape format when dealing with
582 those inodes, which was uncorrectly handled by Linux restore.
583 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
584 providing a test case.
585
586 8. Added the -m parameter to dump which optimises the output for
587 inodes having been changed but not modified since the last dump
588 ('changed' and 'modified' have the meaning defined in stat(2)).
589 For those inodes, dump will save only the metadata, instead of
590 saving the entire inode contents. Inodes which are either
591 directories or have been modified since the last dump are saved
592 in a regular way. Uses of this flag must be consistent, meaning
593 that either every dump in an incremental dump set have the flag,
594 or no one has it.
595
596 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
597 =======================================================================
598
599 1. Fixed behaviour of dump when exceeding resource limits
600 (SIGXFSZ treatment).
601
602 2. Added the -L flag to restore to allow the user to specify a
603 maximal allowed number of miscompares when using restore
604 with the -C option to check the backup.
605
606 3. Detailed the manual entry for the -N option of restore.
607
608 4. Added the -a flag to restore to make able doing unattended
609 restores in -i or -x mode (automatically walks through the
610 multiple dump volumes).
611
612 5. Extended the QFA mode to work with local files and/or
613 remote tapes and files. This way, restore can know in advance
614 the tape number and the offset for the inodes to extract and
615 can minimize the extraction time by seeking directly to the
616 good tape/offset.
617
618 6. Added the -A <archive> option to both dump and restore,
619 which makes dump to archive a dump table-of-contents in
620 the specified file to be used by restore to determine
621 whether a file is in the dump file that is being restored.
622 (the archive file syntax is also compatible with the
623 Solaris ufsdump generated one).
624
625 7. Small fix in restore making it able to read some (broken ?)
626 Solaris ufsdump tapes.
627
628 8. Fixed dump to correctly recognise the root filesystem when using
629 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
630 for reporting this bug.
631
632 9. Added the -P <file> option to restore to create a
633 Quick File Access file from an already made dump. Patch
634 contributed by Uwe Gohlke <uwe@ugsoft.de>.
635
636 10. Made restore compile and run on Solaris, making it a
637 possible replacement for the standard ufsrestore. Port was
638 contributed by Uwe Gohlke <uwe@ugsoft.de>.
639
640 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
641 =====================================================================
642
643 1. Added a set of backup scripts from Eugenio Diaz
644 <getnito@yahoo.com> in the examples section. It features
645 automatic (cron based) full and incremental dumping of
646 several filesystems on a separate filesystem.
647
648 2. Fixed a off-by-one miscalculation which disabled dumping
649 a one letter subdirectory. Thanks to Chris Siebenmann
650 <cks@utcc.utoronto.ca> for reporting the bug.
651
652 3. Fixed several restore bugs occuring when trying to
653 restore some missing files on the tape. Thanks to Chris
654 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
655
656 4. Fixed --with-ldopts configure argument passing, installing from
657 a separate object directory, makefile cleanups contributed
658 by <splite@purdue.edu>.
659
660 5. Fix a bug which could caused, in some conditions, the highest
661 number inode of a filesystem, to not be dumped. Many thanks
662 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
663 chase this bug.
664
665 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
666 =======================================================================
667
668 1. Added a mini howto from Patrick Walsh in the examples
669 directory.
670
671 2. Minor man pages syntax corrections. Thanks to
672 Chris Danis <screechco@home.com> for reporting the bugs.
673
674 3. Added a script from David B. Peterson <dave@toppledwagon.com>
675 to the examples section. It features dumping several
676 filesystems to a remote tape drive upon ssh.
677
678 4. Added a patch provided by Richard Jones <rich@annexia.org>
679 which allows BRADEMAX (number of read errors tolerated by
680 dump) to be adjusted using the -I option.
681
682 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
683 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
684 patch.
685
686 6. Add the -l option to restore to specify if, when doing a
687 remote restore, the file used to access the backup is a
688 regular file (the defaults being a tape drive). Restore needs
689 to know this information when reading a remote compressed
690 dump. Previously, this information was autodetected, but
691 the autodetection code fails (with ioctl: Inappropriate ioctl
692 for device) when using a non Linux remote box. Thanks to
693 many users and especially to Eros Albertazzi
694 <eros@lamel.bo.cnr.it> for reporting this.
695
696 7. Found a workaroung for the dump deadlock problem (3 childs
697 stuck in pause(), father in read()). The workaround seems
698 to work for me and several beta-testers. If it doesn't work
699 for you, please report back.
700
701 8. Updated the RPM spec file (BuildPrereq, URL etc).
702
703 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
704 ========================================================================
705
706 1. Fixed the permissions of a newly created QFA file by dump.
707
708 2. Cleaned up the source of dump (the external variables
709 definition was a complete mess, making possible to have
710 objects overlap).
711
712 3. Fixed restore to use the full tape volume path when doing
713 a compare (since it changes the working directory to the
714 filesystem being compared in the process).
715
716 4. Added the -q option to dump which makes dump abort
717 whenever it needs operator attention. It should be
718 easier to use dump in scripts with this option.
719
720 5. Detect the use of incompatible options to dump and
721 refuse them (like -a and -B options together).
722
723 6. Added bzip2 compression to dump/restore (use option -j level
724 to select it). Note that you will need a restore version
725 >= 0.4b24 in order to restore a bzip2 compressed dump.
726 The same warning as for the zlib compression applies:
727 the tape format of a bzip2 dump is not compatible with the
728 original BSD tape format.
729
730 7. Fixed a overflow problem in dump corrupting the dump when
731 very large files were encountered. Thanks to Vladimir Nadvornik
732 for the bug report.
733
734 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
735 kernel buffer/page cache before starting the dump, helping
736 a bit those who use dump on mounted filesystems. Thanks to
737 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
738 for this suggestion.
739
740 9. Updated the RPM spec file following the RedHat changes
741 (dynamically linked binaries now in /usr/sbin etc).
742
743 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
744 which enables restore to recognise multi volume compressed dumps
745 done on CD. Included his scripts for dump (which pipe the dump
746 output directly into cdrecord) and restore. There is now
747 possible to to backups to CD on the fly!
748
749 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
750 ===================================================================
751
752 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
753 Marcus Meissner (Caldera International Security Dept.).
754
755 2. Implement the Sun rmt extended protocol. Patch submitted
756 by Ian Gordon <iangordon@users.sourceforge.net>.
757
758 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
759 the linux kernel header. This ensures that dump always has the
760 latest version of this file. Patch submitted by
761 Andreas Dilger <adilger@turbolinux.com>.
762
763 4. Report any filesystem present in either /etc/fstab with a
764 positive passno or /etc/dumpdates in dump -w output.
765 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
766
767 5. Fixed the looping problem in dump introduced in the
768 previous version.
769
770 6. Changed the -B option of dump to limit the size of
771 _compressed_ output per volume if compression is on.
772 Patch contributed by Helmut Jarausch
773 <jarausch@igpm.rwth-aachen.de>. Note however that, since
774 it is impossible to predict the size of the compressed
775 data before doing the compression, the -B limit is a bit
776 conservative.
777
778 7. Fixed a bug in reading the operator typed file/tape path for
779 the next volume in restore.
780
781 8. Implemented a "-F script" option to restore which permits the
782 user to specify a script which will be launched at the
783 beginning of each tape, useful for automatic programming of
784 tape changers for example. See the restore man page for the
785 script parameters and return codes.
786
787 9. Small fix for the QFA routines provided by Uwe Gohlke
788 <uwe@ugsoft.de>, and some recommendations for QFA uses in
789 the man pages.
790
791 10. Fixed the multivolume restoring where making a mistake
792 to the 'Mount next tape' prompt caused several blocks to
793 be lost.
794
795 11. Enhanced the -e option of dump to take as a parameter a
796 comma separated list of inode numbers.
797
798 12. Added the -E option to dump which specify a file containing
799 inode numbers to exclude from the dump.
800
801 13. Fixed the compressed multi-volume dump + restore.
802
803 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
804 ==================================================================
805
806 1. Made dump report the number of blocks dumped per volume.
807 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
808
809 2. Fix a bug in dump where waiting too much at the 'change volumes'
810 question caused the volume to be rewritten. Thanks to
811 Graham Reed <greed@users.sourceforge.net> for reporting the
812 bug and providing a patch.
813
814 3. Added a compression option to dump, contributed by Jerry
815 Peters <gapeters@worldnet.att.net>.
816
817 WARNING: the compression works only on files/pipes and
818 drives supporting variable block size.
819
820 WARNING2: tapes created using the compression feature are
821 incompatible with the standard dump tape format, and a
822 version of restore >= 0.4b22 will be needed for extraction.
823
824 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
825 architectures. Thanks to Paul Slootman <paul@debian.org> for
826 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
827 it upstream.
828
829 5. Many cleanups (CPP defines, const char warnings, check of
830 ext2fs COMPAT flags, time_t cleanups, added libext2 version
831 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
832
833 6. Made --prefix option work in configure. All the install path
834 are now based on the configure parameters.
835
836 7. Added the Quick File Access mode in dump/restore, contributed
837 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
838 a file tape position for each inode, and this file is used by
839 restore (if called with parameter Q and the filename)
840 to directly position the tape at the file restore is currently
841 working on. This saves hours when restoring single files from
842 large backups, saves the tapes and the drive's head. Use
843 --enable-qfa option of configure to compile in the QFA support.
844
845 8. Added the possibility to dump several files and directories
846 in a single invocation of dump. Thanks to Uwe Gohlke
847 <uwe@ugsoft.de> for implementing this option.
848
849 9. Fixed the dumping and restoring of files with holes
850 (files ending with holes didn't get dumped or restored
851 correctly).
852
853 10. Fixed a socket descriptor leak in dump, which leaved opened
854 3 file descriptors per dump process (and there is one dump
855 process per tape).
856
857 11. Fixed dump large file system support, by explicit use of
858 open64/lseek64/etc functions (explicit use needed because
859 e2fsprogs libraries don't behave well when compiled with
860 FILE_OFFSET_BITS=64).
861
862 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
863 ======================================================================
864
865 1. Fixed some bugs in the dump -w|-W logic introduced by
866 the previous version. Thanks to Andreas Dilger
867 <adilger@turbolinux.com> for his help on this one.
868
869 2. Fixed again a compilation problem when using old e2fs
870 headers (filesystem label related). Thanks to many users
871 who reported this stupid error.
872
873 3. Fixed a build problem on old lib5 based systems dealing with
874 _PATH_MNTTAB being undefined. Thanks to John Adams
875 <johna@onevista.com> for reporting the bug.
876
877 4. Improved the error detection in restore when creating
878 the temporary files in TMPDIR. Now restore will corectly
879 report a 'no space left on device' error instead of
880 strange errors which could imply an archive corruption.
881 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
882 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
883
884 5. Added the throughput information in dump's progression
885 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
886 for the patch.
887
888 6. Use libext2fs's inode scanning routines, which are particularly
889 robust when dealing with errors (especially when having some
890 bad blocks on the filesystem). Many thanks to Colin
891 <colin@horizon.com> for reporting the bug, investigating
892 the issues, finding a workaround, writing the patch and
893 fully testing it... (of course, if this patch will break
894 anything, he is to blame for :-)).
895
896 7. Made dump and restore LFS aware. Dump can dump a filesystem
897 containing large files, generate a large file on output and
898 restore can restore them. This feature is not enabled by
899 default, use --enable-largefile option of configure to enable
900 it (you need to have a LFS aware glibc though). Thanks to
901 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
902 and to Theodore T'so <tytso@valinux.com> for his always
903 useful thoughts.
904
905 8. Made dump ask upon a tape write error if it should rewrite
906 the current volume (assume this is a bad tape) or if it should
907 assume an end-of-tape condition (useful for tape drives which
908 behaves badly at the end of the tape). Thanks to Andreas
909 Dilger <adilger@turbolinux.com> for the suggestion.
910
911 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
912 =======================================================================
913
914 1. Fixed a small compilation problem due to a change
915 in the definintion of the struct sigaction in
916 glibc 2.0 / libc5. Thanks to Gunther Schlegel
917 <schlegel@riege.de> for reporting the bug and to
918 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
919 a fix.
920
921 2. Modified the label and uuid analysis in order to be
922 self-contained (independant of kernel/libc headers). This
923 should fix the compile with older kernel/libc headers and
924 will preserve the functionality. Thanks to Bernhard Erdmann
925 <bernhard.erdmann@gmx.de> for reporting the bug.
926
927 3. The 'exclude inode' option, if applied to a directory
928 inode, excludes now correctly all the leaves of this
929 directory. Thanks to John R. Dennison
930 <gerdesas@users.sourceforge.net> for reporting the bug.
931
932 4. Fixed the '-e' option to disable the possibility
933 to exclude the root inode (which causes the dumps to
934 be unreadable by restore). Prevented array overflow
935 when multiple -e options are used.
936
937 5. Fixed dump to correctly interpret a filesystem argument
938 which is a mountpoint and it is not an absolute path
939 (as specified in the fstab). Thanks to Bernhard R. Erdmann
940 <be@berdmann.de> for reporting the bug.
941
942 6. Made dump able to backup files larger than 2 GB. Note that
943 dump still doesn't cope with files larger than 4 GB.
944
945 7. Restore the real uid and gid before invoking an external
946 RSH program (big hole when dump or restore is suid root!).
947
948 8. Get the values for the default tape device and the location
949 of fstab file from the system headers. Thanks to
950 Andreas Dilger <adilger@turbolinux.com> for the patch.
951
952 9. Made dump -w|-W report all recognized filesystems
953 present in either /etc/fstab or /etc/dumpdates, and present
954 the list in the same order as in fstab file. Thanks
955 to Andreas Dilger <adilger@turbolinux.com> for the patch.
956
957 10. Made dump's -a (automatic end of tape detection) the
958 default. Specifying one of -B, -c, -d or -s options will
959 override the EOM detection. Thanks to Andreas Dilger
960 <adilger@turbolinux.com> for the patch.
961
962 11. Save the ext2 filesystem label into the dump volume label.
963 Specifying a volume label on the command line (-L option)
964 will override this feature. Thanks to Andreas Dilger
965 <adilger@turbolinux.com> for the patch.
966
967 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
968 =====================================================================
969
970 1. Fixed the signal handling in dump (which I broke in 0.4b17)
971 which was causing several strange dump failures (dump
972 hanged or segmentation faults etc).
973
974 2. Specified the default blocksize in the dump man page.
975
976 3. Changed two info messages of restore to be written on stdout
977 instead of stderr in order to leave stderr only for errors
978 or warnings. Thanks to Stephen Atwell
979 <satwell@urbana.css.mot.com> for the suggestion.
980
981 4. Corrected an off by one calculation which prevented
982 dumping a directory having a 1 character name length.
983 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
984 for reporting the bug.
985
986 5. Reinforce dump in order to allow only level 0 and no
987 -u option when dumping a subdirectory, and document
988 this in the man page. Thanks to Bernhard Erdmann
989 <bernhard.erdmann@gmx.de> for reporting the bug.
990
991 6. Fixed a small harmless bug in restore which caused
992 in some conditions hard links to be restored several
993 times (therefore generation some warning messages).
994 Thanks to Kenneth Porter <shiva@well.com> for
995 reporting the bug.
996
997 7. Updated the RPM spec file to the latest RedHat version,
998 providing FHS packaging and other cosmetic changes.
999 You will need RPM version >= 3.0.5 in order to build the RPM.
1000
1001 8. Updated the configure script to check for libtermcap
1002 before checking for libreadline, because we need this
1003 library in order to compile the readline support.
1004
1005 9. Made dump understand the LABEL= and UUID= notation
1006 both in /etc/fstab and on the command line. Note that
1007 you will need the /proc filesystem in order to use
1008 these notations. Thanks to Erik Troan <ewt@redhat.com>
1009 for providing the patch.
1010
1011 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
1012 ===================================================================
1013
1014 1. Fixed a potential buffer overflow in restore. Thanks
1015 to Stan Bubrouski <satan@fastdial.net> for reporting
1016 the bug.
1017
1018 2. Fixed a readline-related bug which prevented
1019 'cat DUMPFILE | restore -i -f -' from working. Thanks
1020 to Charles Karney <karney@users.sourceforge.net>
1021 for the bug report.
1022
1023 3. Changed a few "panic" into "exit", causing restore to
1024 be more stable against some attacks (like the last one
1025 reported on Bugtraq, although the last version of restore
1026 was not vulnerable - just dumped core). Thanks to
1027 Andreas Hasenack <andreas@conectiva.com.br> for reporting
1028 the bugs to me.
1029
1030 4. Removed the suid-root bit on dump and restore in the
1031 default build (and generated RPMs). It should be safer
1032 now. Users who need the suid-root capabilities in order
1033 to make network backups should read first the man page
1034 of dump and enable the suid bit manually.
1035
1036 5. Added -ltermcap to the compile parameters for restore
1037 when using readline support, in order to make the compile
1038 process work on some systems (SuSE ?). Thanks to
1039 Patrik Schindler <poc@pocnet.net> for reporting the bug.
1040
1041 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
1042 ====================================================================
1043
1044 1. The -F script is called now *only* at the end of a tape,
1045 not at the end of the dump. Thanks to Christian Haul
1046 <haul@informatik.tu-darmstadt.de> for the bug report.
1047
1048 Normally, the device name and the current volume number
1049 are passed on the command line of the script. However,
1050 if you want the old style script (with no arguments
1051 passed on the command line), you can enable it in
1052 configure with the --enable-oldstylefscript.
1053
1054 2. Use posix signal handling to preserve dump functionality
1055 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
1056 the patch.
1057
1058 3. Made the exit code of restore in case of a 'C'ompare
1059 command reflect the result of the compare. An exit status
1060 of 0 means the dump archive is correct, 1 means tape errors,
1061 2 means that some files were modified. Thanks to Kenneth Porter
1062 <shiva@well.com> for the suggestion.
1063
1064 4. Made (finally) quotes work in interactive restore.
1065
1066 5. Small fixes in order to allow dump to compile with a
1067 really old e2fsprogs version. Thanks to Ian Zimmerman
1068 <itz@speakeasy.org> for the bug report.
1069
1070 6. Add GNU readline capabilities to interactive restore.
1071 Use configure's --enable-readline flag to enable this feature.
1072 Thanks to Patrik Schindler <poc@pocnet.net> for the
1073 suggestion.
1074
1075 7. Do the compare on the fly in restore 'C' mode (this will
1076 allow not to exhaust the available /tmp space when
1077 ccmparing large files). Thanks to Kenneth Porter
1078 <shiva@well.com> for the suggestion.
1079
1080 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
1081 ====================================================================
1082
1083 1. Fixed some several duplicate 'DUMP: DUMP:' in the
1084 output of dump.
1085
1086 2. Corrected the estimation of blocks to dump. Note that
1087 this is still wrong for large directory inodes, due
1088 to the size of a BSD directory entry on the tape
1089 (estimation is based on the size of the ext2 directory,
1090 which is smaller because it doesn't contain the null
1091 character at the end).
1092
1093 3. Made dump report the total number of MB written to
1094 tape at the end of run. Thanks to W. Reilly Cooley
1095 <wcooley@nakedape.cc> for the patch.
1096
1097 4. Added the -X option to restore giving the possibility
1098 to read the names of the files to be extracted/listed
1099 from a text file (in addition of the command line).
1100 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
1101 patch.
1102
1103 5. Added the device name and the current volume number
1104 as arguments to the end of tape script (dump -F option).
1105
1106 6. Made the multi-volume dump work again (after having
1107 broken it in 0.4b15).
1108
1109 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
1110 ===================================================================
1111
1112 1. Added a prompt command in interactive restore mode. Thanks
1113 to Andreas Dilger <adilger@home.com> for the patch.
1114
1115 2. Fixed a buffer overflow problem in dump (caused by
1116 not checking the size of the filesystem parameter).
1117 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
1118 reporting this on Bugtraq (and to several dump users
1119 who forwarded me his mail).
1120
1121 3. Added the '-F script' option to dump in order to
1122 launch a script at the end of each tape (to be used
1123 with a tape changer, or to notify the sysadmin by
1124 pager etc.).
1125
1126 4. Fixed a bug in restore compare code caused by the changes
1127 I made in 0.4b14.
1128
1129 5. Fixed the treatment of options using the old BSD syntax
1130 in both dump and restore.
1131
1132 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
1133 =======================================================================
1134
1135 1. Fixed a bug in dump which may have caused invalid deleted
1136 directories to be dumped out if they were deleted after the
1137 mapping pass. This could occure on active filesystem and lead
1138 to heap corruption (causing dump malfunction in many possible ways).
1139 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
1140 bug and submitting the patch.
1141
1142 2. Fixed the handling of the filesystem parameter in dump. A
1143 '/mnt/disk' parameter caused the disk contents to be dumped,
1144 but a '/mnt/disk/' parameter caused the mountpoint directory
1145 to be dumped (generally an empty directory).
1146
1147 3. Improved the output of dump in order to tell which directory
1148 it is currently dumping (when dumping a subtree).
1149
1150 4. Added the '-e' exclude inode option to dump. Thanks to
1151 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
1152
1153 5. Added a REPORTING-BUGS file in order to provide a guide
1154 on how to correctly report a bug in dump/restore.
1155
1156 6. Fixed a restore problem when restoring a hard link to an inode
1157 having the immutable or append-only attribute set. Thanks to
1158 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
1159
1160 7. Fixed a compatibility problem between dumps made on little
1161 endian machines (the format was correct) and big endian
1162 machines (the format was incorrect). This fix break the
1163 compatibility with the older dumps made on big endian
1164 machines (sparc, m86k, ppc etc). For the first time in
1165 linux dump's history (I believe), the dumps made by *BSD,
1166 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
1167 so interchangeable. Thanks to Rob Cermak
1168 <cermak@ahab.rutgers.edu> for submitting the bug and
1169 helping me test the fix.
1170
1171 8. Fixed the way dump reports the remaining percent/time, if
1172 the number of blocks actually dumped exceeds the estimated
1173 number of blocks. Thanks to Jean-Paul van der Jagt
1174 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
1175
1176 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
1177 ======================================================================
1178
1179 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
1180 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
1181
1182 2. Removed the requirement to build the RPM as root from the
1183 spec file. Thanks to Christian Weisgerber
1184 <naddy@mips.rhein-neckar.de> for submitting this.
1185
1186 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
1187 causing dump to block when dumping really huge filesystems.
1188 Many thanks to Patrik Schindler <poc@pocnet.net> for
1189 helping me find this bug.
1190
1191 4. Fixed the treatment for an interrupt signal when dump access
1192 the remote tape through RSH. Thanks to Christian Weisgerber
1193 <naddy@mips.rhein-neckar.de> for providing the patch.
1194
1195 5. Fixed a bug which was causing dump/restore to display
1196 garbage characters instead of the remote host name.
1197
1198 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
1199 =====================================================================
1200
1201 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
1202 <kukuk@suse.de> for submitting the patch.
1203
1204 2. Fix for the exit code when using the size estimate option of
1205 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
1206 submitting the patch.
1207
1208 3. Handle EINTR in atomical reads/writes in dump, which was causing
1209 dump to fail on some systems. Thanks to Eric Jergensen
1210 <eric@dvns.com> for reporting the bug and submitting the patch.
1211
1212 4. Handle more than 16 characters for the device names in dumpdates.
1213 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
1214 tracking down the problem and proposing the solution.
1215
1216 5. Fixed a bug in dump which prevented the creation of the
1217 dumpdates file when doing a 0-level dump without already
1218 having a dumpdates file. Thanks to Patrik Schindler
1219 <poc@pocnet.net> for reporting the bug.
1220
1221 6. Changed the way dump 'S' flag reports the size estimate
1222 from number of blocks into bytes (making it compatible
1223 with the Solaris version, and simplifying things for
1224 amanda users). Thanks to Jason L Tibbitts III
1225 <tibbs@math.uh.edu> for reporting the bug.
1226
1227 7. Fixed a compatibility problem in linux/alpha dump tape format.
1228 Now the linux/alpha dump are (again) compatible with the
1229 other dump formats. But this breaks compatibility with
1230 older dumps made on alpha. Thanks to Mike Tibor
1231 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
1232
1233 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
1234 ======================================================================
1235
1236 1. Added a '--enable-kerberos' to configure.
1237
1238 2. Added a 'S' option to dump which determines the amount of space
1239 that is needed to perform the dump without actually doing it, similar
1240 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
1241 <cermak@ahab.rutgers.edu>.
1242
1243 3. Added a 'M' multi-volume option to dump and restore which enables
1244 dumping to multiple files (useful when dumping to an ext2
1245 partition to make several dump files in order to bypass the 2GB
1246 file size limitation). The 'f' argument is treated as a prefix and
1247 the output files will be named <prefix>001, <prefix>002 etc. With
1248 the 'M' flag, restore automatically selects the right file without
1249 asking to enter a new tape each time.
1250
1251 4. Fixed a memory leak which was causing dump to grow very big
1252 (270MB when dumping a 10GB filesystem...). Thanks to Jason
1253 Fearon <jasonf@netrider.org.au> for reporting the bug.
1254
1255 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
1256 ======================================================================
1257
1258 1. Make configure test if the system glob routines support
1259 extended syntax (ALTDIRFUNC). If not, use the internal glob
1260 routines instead of system ones. Thanks to Bernhard Sadlowski
1261 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
1262 and helping me resolve this and other minor libc5 compiling
1263 glitches.
1264
1265 2. Fix a problem when dumping a ext2fs with the 'filetype'
1266 feature enabled. Thanks to Patrick J. LoPresti
1267 <patl@cag.lcs.mit.edu> for reporting the bug and to
1268 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
1269
1270 3. Made the nodump flag work on directories. A directory which
1271 has the nodump flag gets never dumped, regardless of its
1272 contents.
1273
1274 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
1275 which allows dump on an active ext3 filesystem. However, this
1276 is a "quick and dirty" patch which enables backup of an ext3
1277 filesystem through the ext2 compatibility (by ignoring the
1278 NEEDS_RECOVERY bit). The journal file is not recognized and
1279 it is dumped (it should not).
1280
1281 5. Test the superblock compatibility flags when dumping, in order
1282 to be sure that we know how to deal with specific features.
1283
1284 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
1285 ====================================================================
1286
1287 1. Use lchown instead of chown, fixing a possible security problem
1288 when restoring symlinks (a malicious user could use this
1289 to deliberately corrupt the ownership of important system files).
1290 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
1291 this and providing the patch.
1292
1293 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
1294 ====================================================================
1295
1296 1. Put dump sources under CVS, added Id tags in all files so
1297 one can use 'ident' on binary files.
1298
1299 2. Added the dump/restore version in the usage text so one can
1300 easily verify the version he is using.
1301
1302 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
1303 a va_start/va_end problem on linux-ppc (always call va_start
1304 va_end in pairs each time we use a vararg function).
1305
1306 4. Added again the DT_* constants because old libc does not
1307 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
1308 for submitting the bug report.
1309
1310 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
1311 this should enable dumping big (huge) filesystems.
1312
1313 6. Added the RSH environment variable in order to be able to
1314 use a rsh replacement like ssh when doing remote backups (and
1315 bypass the security limitations of rcmd). Now you can do remote
1316 backups without being root (or making dump setuid root).
1317
1318 7. Modified again the way dumpdates works. For incremental dumps,
1319 we need to read dumpdates even if we are not using 'u' option.
1320 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
1321 this should work.
1322
1323 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
1324 ===================================================================
1325
1326 1. Removed the 'k' flag from the restore 'about' text if kerberos
1327 was not compiled in.
1328
1329 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
1330 to them (fsetflags takes a char*, setflags an open fd!).
1331
1332 3. (f)setflags is called only if the flags aren't empty. If the
1333 file is a special file, a warning is printed, because changing
1334 flags implies opening the device. Normally, a special file
1335 should not have any flag... (Debian bug #29775, patch provided
1336 by Abhijit Dasgupta <abhijit@ans.net>).
1337
1338 4. Made possible to dump a file system not mentioned in /etc/fstab.
1339 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
1340
1341 5. Changed the default behaviour to not create dumpdates
1342 unless 'u' option is specified. Removed the old "debian-patch"
1343 which provided the same thing. (Debian bug #38136, #33818).
1344
1345 6. Removed all those dump*announce, since they were getting old...
1346
1347 7. Added warning messages if dumpdates does not exist and
1348 when an inferior level dump does not exist (except for a level 0
1349 dump).
1350
1351 8. Debugged the glob calls in interactive mode: restore used a
1352 dirent struct which was different from the /usr/include/dirent.h
1353 one (this used to work, is it a glibc2 change?), so none of the
1354 compat glob (which used /usr/include/dirent.h) or the system glob
1355 worked. Restore use now the system dirent (and the system
1356 DT_* constants), which are compatible with BSD ones.
1357
1358 9. Added a configure flag (--with-dumpdatespath) to specify
1359 the location of dumpdates file. By default, it is
1360 /etc/dumpdates.
1361
1362 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1363 included the current date/version in man pages.
1364
1365 11. Corrected the estimation of remaining time when
1366 the operator doesn't change the tapes quickly enough. This
1367 was an old bug, I thought I corrected it, and discovered
1368 that in fact it was corrected in two different places, so
1369 the results canceled each other...
1370
1371 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1372 ===================================================================
1373
1374 1. Integrated multiple patches from RedHat, Debian and SuSE:
1375
1376 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1377 isn't specified.
1378 - several fixes in the man pages.
1379 - update the default tape device to /dev/st0.
1380 - many updates for Linux Alpha (byte ordering, size_t etc).
1381 - buffer overruns.
1382 - use environment variable for TMPDIR (instead of /tmp).
1383 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1384 - workaround egcs bug (RedHat bugs #4281 and #2989).
1385 - wire $(OPT) throughout Makefile's.
1386
1387 2. Upgrade the dump revision to 1, making possible to dump filesystems
1388 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1389
1390 3. Fix some compile warnings, prototype all functions.
1391
1392 4. Use glibc err/glob instead of internal compatibility
1393 routines (only if available).
1394
1395 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1396 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1397
1398 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1399 ======================================================================
1400
1401 1. Integrated the changes from FreeBSD-3.1-RELEASE
1402 (mostly bug fixes, buffer overruns, dump has now an "automatic
1403 tape length calculation" flag, dump/restore can use kerberos now
1404 (this is NOT tested), use environment variables for TAPE and
1405 RMT etc.).
1406
1407 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1408
1409 3. Corrected a bug in restore when using 'C' option with multi-volumes
1410 tapes (files splited accros two tapes give "size changed" errors
1411 when comparing).
1412
1413 4. Corrected the long standing bug when dumping multiple tapes.
1414 This works for me, needs further testing.
1415
1416 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1417 ====================================================================
1418
1419 1. Dump now runs correctly on kernels 2.1.x
1420 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1421
1422 Changes between versions 0.4b2 and 0.4b3
1423 ========================================
1424
1425 1. Use realpath() if available
1426
1427 2. Report statistics
1428
1429 Changes between versions 0.4b1 and 0.4b2
1430 ========================================
1431
1432 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1433 the patch)
1434
1435 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1436
1437 3. Fixed configure.in to correctly handle --enable-rmt
1438
1439 Changes between versions 0.3 and 0.4b1
1440 ======================================
1441
1442 1. Integrated the changes from 4.4BSD-Lite2
1443
1444 2. Integrated the patches from Debian and Red Hat
1445
1446 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1447
1448 4. Changed dump to use the Ext2fs library to get block addresses. This
1449 should solve the endianness problem on SparcLinux.
1450
1451 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1452 distribution's one) to use autoconf
1453
1454 6. Fixed a few minor bugs
1455
1456 Changes between versions 0.2e and 0.2f
1457 ======================================
1458
1459 1. Added the creation of named pipes (fifos) in restore.
1460
1461 2. Added the -N flag in the restore manual page.
1462
1463 3. Added the file kernel.patch which contains the llseek() optimization
1464 patch for 1.2.x kernels.
1465
1466 4. Fixed a bug in the restoration of symbolic links: owner and group were
1467 not restored.
1468
1469 5. Integrated some changes from FreeBSD 2.2.
1470
1471 6. Added a call to ftruncate() after restoring each file to restore
1472 correctly files ending by a hole.
1473
1474 Changes between versions 0.2d and 0.2e
1475 ======================================
1476
1477 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1478 with owner = root (0) and group = root/wheel/whatever (0).
1479
1480 Changes between versions 0.2c and 0.2d
1481 ======================================
1482
1483 1. Dump is now able to backup 2GB+ filesystems.
1484
1485 2. Dump and restore can now be linked as static binaries.
1486
1487 Changes between versions 0.2b and 0.2c
1488 ======================================
1489
1490 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1491
1492 Changes between versions 0.2a and 0.2b
1493 ======================================
1494
1495 1. Really fixed the bug that I should have corrected in 0.2a.
1496
1497 2. Enabled optimization again.
1498
1499 Changes between versions 0.2 and 0.2a
1500 =====================================
1501
1502 1. Disabled the optimization during compilation.
1503
1504 Changes between versions 0.1 and 0.2
1505 ====================================
1506
1507 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1508 the fs_type field (actually, I modified the file fstab.c to make it
1509 use the mntent functions).
1510
1511 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1512 made by dump should be readable by the BSD restore and backups made
1513 by the BSD dump should be readable by restore. Unfortunately, this
1514 means that the dump format has changed between version 0.1 and version
1515 0.2 :-(
1516
1517 3. Dump is now able to backup a subtree, it is no longer limited to whole
1518 filesystems like the BSD version.
1519
1520 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1521 than 2 GB.
1522
1523 Changes between versions 0.0 and 0.1
1524 ====================================
1525
1526 1. Now create links rdump and rrestore during the `make install' step.
1527
1528 2. Linux port specific bugs added to the manual pages
1529
1530 3. Incorrect estimation of the number of tapes blocks fixed when doing
1531 incremental backups.
1532
1533 4. Better ls-like format in restore in interactive mode.