Fix fgets bug when asking for a tape name.
[dump.git] / CHANGES
1 $Id: CHANGES,v 1.326 2011/06/08 15:40:53 stelian Exp $
2
3
4 Changes between versions 0.4b43 and 0.4b44 (released ?????????????)
5 ===================================================================
6
7 1. Recognize ext4 partitions in dump -w commands (Sourceforge
8 bug #3125587, RedHat bug #658890). Thanks to Jan Görig
9 <jgorig@users.sourceforge.net> for reporting the bug and
10 submitting the patch.
11
12 2. Fix an issue with multi volume backups, which breaks
13 restoring of files which are splitted on two or more
14 volumes and are starting on 2nd tape or later (RedHat
15 bug #507948).
16
17 3. Fix a bug in dump when dumping files with holes. This bug
18 was introduced in 0.4b42 while adding ext4 support.
19 (Sourceforge bug #3133762).
20
21 4. Fix a bug in restore -P where useless index files for compressed
22 dumps were created (Debian bug #583093). Thanks to
23 Mark Wooding <mdw@distorted.org.uk> for the patch.
24
25 5. Fix SYS_clone invocation on s390 architectures (Sourceforge
26 bug #3303235). Thanks to Adam Tkac <atkac@redhat.com> for
27 reporting the bug.
28
29 6. Fix static builds. Thanks to Peter Volkov <pva@gentoo.org>
30 for the patches.
31
32 7. Fix false negatives in configure zlib test. Thanks to
33 Bear Giles <bgiles@coyotesong.com> for the patch.
34
35 8. Add quick regression script from Bear Giles <bgiles@coyotesong.com>.
36
37 9. Fix a bug introduced in 0.4b43 causing restore to crash when
38 asking for a new volume when doing multi-volume restores. Thanks
39 to Patrik Schindler <poc@pocnet.net> for reporting the bug.
40
41 Changes between versions 0.4b42 and 0.4b43 (released June 11, 2010)
42 ===================================================================
43
44 1. Fix a bug in dump making impossible to handle large toc files
45 (> 2 GB). Thanks to X DUGi <xdugi@users.sourceforge.net> for
46 reporting the bug (Sourceforge bug #2820629)
47
48 2. Fix 'restore -x' in multi-volume mode, which caused files being
49 spanned on the first and second volumes to be incorrectly extracted.
50 Thanks to Adam Tkac <atkac@redhat.com> for reporting the bug.
51 (there may be several bug reports about this issue, including
52 Sourceforge bug #2117008, RedHat bugs #232415 and #444958).
53
54 3. Remove -lselinux -lsepol from the standard libraries list,
55 they shouldn't be needed if selinux is not enabled in dump
56 (this also fixes a build failure when selinux libraries are
57 not installed).
58
59 4. Fix restoration of extended attributes of fifos and device nodes.
60 Thanks to Igor Zhbanov <izh1979@gmail.com> for the bug report.
61
62 5. Don't attempt to set extended attributes on files that were not
63 extracted in 'restore -N' mode. Thanks to Jan Görig
64 <jgorig@users.sourceforge.net> for the patch.
65
66 6. Make the dump man page explicit about ext4 support.
67
68 7. Fix a bug in dump which caused EA entries to be unrecognizable by
69 restore in some corner cases. Add a workaround to restore making
70 it able to properly deal with those corrupted EA entries. Thanks
71 to John Austin <jaustin1@users.sourceforge.net> for reporting
72 the bug.
73
74 8. Fix some issues when restoring a dump which was generated using
75 exclusion patterns (either via -e or via the nodump attribute)
76 (see the Debian bug #574667 for details. Thanks to Frédéric
77 Brière <fbriere@fbriere.net> for the bug report and the associated
78 patch.
79
80 9. Fix progress/estimated blocks display in dump when doing really
81 huge backups. Thanks to Steve Bonds <sbonds@users.sourceforge.net>
82 for the bug report and tests. (Sourceforge bug #2987758)
83
84 10. Improve level 1 dump speed by rearanging the "mapdirs" code.
85 Thanks to Andreas Kies <andikies@t-online.de> for finding the
86 bottleneck and sending a patch (Sourceforge bug #2998119).
87
88 11. Use only ctime in order to compare newness instead of both
89 ctime and mtime. This should both speed up dumping time and
90 correct some side-effect bugs (like dumping files with an mtime
91 in the future in every incremental backup). Thanks to
92 Kieran Clancy <codebeard@users.sourceforge.net> for reporting the
93 bug (Sourceforge bug #2999207).
94
95 12. Extract dumped UNIX sockets instead of ignoring them.
96 (Sourceforge bug #3007216).
97
98 13. Compiler warning fixes (mainly warn_unused_result ones).
99
100 Changes between versions 0.4b41 and 0.4b42 (released June 18, 2009)
101 ===================================================================
102
103 1. Fix printout of the 'Connection to' message in dump/restore.
104 Thanks to Dale Wiles <dwiles@users.sourceforge.net> for the
105 bug report and its fix.
106
107 2. Fix dump -w/-W output for filesystems having the last backup
108 level equal to 0. Thanks to Pascal Bouchareine
109 <kalou@users.sourceforge.net> for reporting the bug.
110
111 3. Enable restore to handle restoring onto a different SELinux
112 policy from the dump. Thanks to Tony Nelson
113 <tony_nelson@users.sourceforge.net> for the patch.
114 (see http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=189845
115 for details)
116
117 4. Fix EA set failures when restoring immutable files. Thanks
118 to Andrew Kroeger for the patch.
119
120 5. Fix "mode file too large" errors when restoring huge backups.
121 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
122 report.
123
124 6. Add the ability to link against libtinfo library if present,
125 instead of libtermcap or libncurses libraries. Thanks to
126 Adam Tkac <atkac@redhat.com> for the patch.
127
128 7. When comparing a backup, do not consider to be an error the fact
129 that we encounter files created while dump was in progress.
130 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
131 report.
132
133 8. Force '-a' to be the default value when running 'restore -C'.
134 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
135 report.
136
137 9. Many improvements to the 'cron_dump_to_disk' example.
138 Thanks to Aaron S. Hawley <ashawley@users.sourceforge.net> for
139 the patch.
140
141 10. Fix the QFA generation when extended attributes are backuped
142 along with the inodes. Prior to this fix, the entries for
143 some inodes may get corrupted (duplicated entries or incorrect
144 ones), making the QFA file unusable for repositionning in
145 restore. Many thanks to Kenneth Porter <shiva@sewingwitch.com>
146 for the bug report and the testing of the fix.
147
148 11. Fix EA restoration in debug mode (Sourceforge bug #1986216).
149
150 12. Stefan Auracher <star7@onlinehome.de> noticed that the Tower
151 of Hanoi backup strategy documented in the man page wasn't
152 actually based on the Tower of Hanoi algorithm. The man page
153 was updated accordingly.
154
155 13. Fix missing level information from dump output when using
156 the default level (RedHat bug #493635). Thanks to Adam Tkac
157 <atkac@redhat.com> for the patch.
158
159 14. Fix the dump man page to reflect the fact that the default
160 dump level is 0 not 9 (RedHat bug #356121). Thanks to Adam Tkac
161 <atkac@redhat.com> for the patch.
162
163 15. Fixed a few spelling errors in the man pages (RedHat bug #489853).
164 Thanks to Adam Tkac <atkac@redhat.com> for the patch.
165
166 16. Fixed restore man page and program usage which incorrectly stated
167 that -P and -A were both allowed in the same invocation (RedHat
168 bug #490627). Thanks to Adam Tkac <atkac@redhat.com> for the patch.
169
170 17. Use sys_clone under Linux to share I/O contexts between dump
171 processes, thus drastically increasing the performance of dump
172 under CFQ (which is the default I/O scheduler used in a number of
173 distributions). Thanks to Jeff Moyer <jmoyer@redhat.com> for the
174 patch.
175
176 18. Add (preliminary) ext4 support - thanks to libext2fs which does
177 all the job for us. Thanks to Gertjan van Wingerde
178 <gwingerde@gmail.com> for the patch.
179
180 Changes between versions 0.4b40 and 0.4b41 (released January 2, 2006)
181 =====================================================================
182
183 1. Fix restore of dumped Access Control Lists. The previous
184 code for EA works fine, but ACLs needed conversion from the
185 ext2/3 disk format to posix_acl format before restoring.
186
187 2. Fix some issues with restoration of EA on big endian
188 platforms.
189
190 3. Fix restore when the symtab is over 2GB in size.
191
192 4. Made the directory hash indexing an optional feature,
193 accessible by the '-H' option of restore, and disable it
194 by default.
195
196 5. Fixed dump to not include extended attributes information
197 in the toc (archive) file which confused restore -t.
198
199 Changes between versions 0.4b39 and 0.4b40 (released May 2, 2005)
200 =================================================================
201
202 1. Changed restore to emit warnings (instead of emitting a fatal
203 error) if a file (or a directory) is unavailable for a
204 comparision (if the user doesn't have the necessary permissions
205 to access it for example). Thanks to Kenneth Porter
206 <shiva@sewingwitch.com> for the bug report.
207
208 2. Re-done the 'do not save directory entries to non-dumped inodes
209 (excluded from dump)' feature. The previous implementation
210 worked well for excluded directories but not for regular files.
211 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
212 report.
213
214 3. Fixed a bug in dump where the tape size was miscalculated when
215 the user used -d/-s to specify the tape characteristics. Thanks
216 to Philip Goisman <goisman@physics.arizona.edu> for reporting
217 the bug.
218
219 4. Fixed another bug introduced in restore with the hashtree
220 implementation. This one caused restore to stop saying
221 "removenode: non-empty directory" in some cases.
222
223 5. Added support for dumping and restoring ext2/3 extended
224 attributes (EA), like the access control lists (ACL) or
225 the security labels used by SELinux.
226
227 Changes between versions 0.4b38 and 0.4b39 (released January 21, 2005)
228 ======================================================================
229
230 1. The newly added dump_on_cd_3 example was buggy, replace it
231 with an updated version from Andrew Basterfield
232 <bob@cemetery.homeunix.org>.
233
234 2. Made restore to chdir() back into the initial directory when
235 dumping core while aborting a comparision operation. The
236 previous behaviour was to write the corefile at the root of
237 the directory being compared, which could very well be
238 read only and preventing the corefile generation. Thanks
239 to Kenneth Porter <shiva@sewingwitch.com> for the bug report.
240
241 3. Silenced the failure to call fgetflags() when comparing an
242 entry which has no ext2 attributes (as in lsattr()).
243
244 4. Fix a brown paper bug in restore -C which broke restore and
245 caused modifications on the filesystem being compared
246 (directories containing a file with the same name as the
247 directory get renamed to RSTTMP...). Thanks to Kenneth Porter
248 <shiva@sewingwitch.com> for finding the bug and helping me
249 reproduce it.
250
251 5. Made restore -C force the -N flag (no writing allowed on
252 the disk) in order to prevent more bugs like the above one.
253
254 Changes between versions 0.4b37 and 0.4b38 (released January 7, 2005)
255 =====================================================================
256
257 1. Fix a couple of troff syntax bugs in the man pages.
258 Thanks to Eric Raymond <esr@thyrsus.com> for the patch.
259
260 2. Made restore use either libncurses or libtermcap, depending
261 on which one is available at configure time.
262
263 3. Fixed restore negative size display bug when comparing a
264 dump containing files over 2GB. Thanks to Steve Bonds
265 <sbonds@users.sourceforge.net> for the bug report.
266
267 4. Do not save directory entries to non-dumped inodes
268 (excluded from dump). This will eliminate the 'missing
269 file' warnings when doing 'restore -C'.
270
271 5. Fix dump crash when backuping a huge (2TB) filesystem,
272 due to a bogus calculation on the inode map size.
273 Thanks to Kevin B. Haines <K.B.Haines@rl.ac.uk> for
274 submitting the bug and testing the fix.
275
276 6. Fix a problem in restore where the final \0 in the symbolic
277 link names could have been lost, generating corrupt filenames.
278 Thanks to Kyle Wilson <kyle.wilson@amd.com> for reporting the
279 bug.
280
281 7. Implemented a hash list for the directory names in restore.
282 The linear list used before caused problems in interactive
283 restores when dealing with directories having thousands of
284 entries. Thanks to Brian Ristuccia <bristuccia@starentnetworks.com>
285 for reporting the bug.
286
287 8. Improved restore -C, this time including the directory
288 attributes into the comparision.
289
290 9. Made restore understand tapes containing EA/ACLs (which will
291 be dumped by the next version of dump). In this version
292 extended attributes on the tape are ignored, for full EA/ACL
293 support wait for the next version or try the experimental EA
294 patch.
295
296 Changes between versions 0.4b36 and 0.4b37 (released July 7, 2004)
297 ==================================================================
298
299 1. Added the --enable-staticz configure option which enables
300 dump and restore to be linked with static versions of
301 libz and libbz2 (and dynamic versions of all other libraries).
302 This will make Debian users happy, because libz and libbz2
303 were the only needed libraries living in /usr, all the
304 others live in /lib. In case of system emergency, it is
305 better not to have to rely on an extra filesystem.
306 Thanks to Bdale Garbee <bdale@gag.com> for the suggestion.
307
308 2. Fix compilation on (at least the Linux Debian port to) AMD64.
309 (<ext2fs/ext2_types.h> defines some types (__s64 and __u64)
310 that are also defined by <linux/types.h> (<asm/types.h>) and
311 they conflict).
312
313 3. Make dump's reading of the dumpdates file a bit more robust,
314 preventing dump from crashing when the dumpdates file has
315 been modified by hand.
316
317 4. Fixed some offset calculations in dump code which could
318 lead to "bread lseek errors" on large filesystems. Thanks
319 to Bruce Lowekamp <lowekamp@users.sourceforge.net> for
320 reporting this bug and debugging the issue.
321
322 5. Made dump use the blkid library when searching for devices
323 by label or uuid instead of dump's own routines.
324
325 6. Corrected a bug in dump where a wrong LABEL=... line in
326 /etc/fstab could prevent dump from dumping unrelated
327 filesystems. Thanks to Bruce Lowekamp
328 <lowekamp@users.sourceforge.net> for reporting the bug.
329
330 Changes between versions 0.4b35 and 0.4b36 (released April 21, 2004)
331 ====================================================================
332
333 1. Fixed dump compilation with old gcc versions. Thanks to
334 Mike Castle <dalgoda@users.sourceforge.net> for the patch.
335
336 2. Fixed some warnings (howmany, roundup, powerof2 redefined)
337 when compiling against a recent glibc version.
338
339 3. Fixed a bug in restore preventing the read of a dump tape
340 written with Solaris 7 ufsdump. Thanks to Patrick Higgins
341 <phiggins@transzap.com> for reporting the bug and providing
342 the test case.
343
344 4. Changed dump to enable the creation of volumes of different
345 sizes in a single run (make -B accept a list of values).
346 Patch contributed by Florian Zumbiehl <florz@gmx.de>.
347
348 5. Use the glibc provided minor() and major() macros instead
349 of our own bitmask implementation. This should be safe for
350 when the major/minor namespace will migrate to 32 bits.
351 Thanks to Zhang Jun <zhangjun@nanjing-fnst.com> for reporting
352 the bug.
353
354 6. Made explicit in the dump man page that dump will not create
355 a remote file, it will only write to an already existing one.
356
357 7. Another try at making size estimates better again.
358
359 8. Put back the inconditional running of the end-of-tape
360 script which was changed in 0.434 to be run only when -M
361 or multiple -f were NOT specified. Some users rely on this
362 feature even when it is combined with -M/-f.
363
364 9. Fixed restore when restoring huge backups (where rstdir...
365 temporary files are over 2GB). Thanks to
366 Raphael Jaffey <rjaffey@artic.edu> for reporting this,
367 debugging the issue and making the patch.
368
369 10. Made restore understand FreeBSD UFS2 tapes. Thanks to
370 David <vrtin@users.sourceforge.net> for submitting the bug
371 and providing a test case.
372
373 11. Made dump work with an arbitrary integer level (no more 10
374 levels only limitation). Thanks to Matthew
375 <msvincen@midway.uchicago.edu> for the patch.
376
377 Changes between versions 0.4b34 and 0.4b35 (released December 21, 2003)
378 =======================================================================
379
380 1. Added a note in the dump man page saying that the default
381 blocksize can be 32 if -d is used with a high density argument.
382 Thanks to Antonios Christofides <A.Christofides@itia.ntua.gr>
383 for the patch.
384
385 2. Fixed configure to correctly understand CPPFLAGS, CFLAGS,
386 LDFLAGS environment variables. Thanks to Arcady Genkin
387 <antipode@users.sourceforge.net> for reporting the bug.
388
389 3. Made -e/-E options of dump accept an unlimited number of inodes
390 to be excluded, rather than a hardcoded maximum. Thanks to
391 Dietrich Rothe <d-rothe@users.sourceforge.net> for the patch.
392
393 4. Updated the autoconf system to 2.50. Forced the -D_BSD_SOURCE
394 and -D_USE_BSD_SIGNAL defines in configure in order to solve
395 64bit build problems because quad_t is redifined with a
396 different signature. Thanks to Mike Harris <mharris@redhat.com>
397 for reporting this bug.
398
399 5. Made restore build on Solaris, making possible to
400 restore Linux's "enhanced" tapes. Thanks to Uwe Gohlke
401 <uwe@ugsoft.de> for the patch.
402
403 6. Made an extension in the dump tape format capable of saving
404 MacOSX specific inode extensions. Uwe Gohlke <uwe@ugsoft.de>
405 wrote the extension and contributed the restore code back
406 into this codebase. The same extension mechanism will be
407 used in the future to save ACLs...
408
409 7. Made rmt work correctly with regard to QFA and local/remote
410 files and tapes. The remote access will however work only
411 when the dump provided rmt version is used. If you want to
412 use another rmt server, please do not use the QFA feature.
413 Thanks to Uwe Gohlke <uwe@ugsoft.de> for the patch.
414
415 Changes between versions 0.4b33 and 0.4b34 (released April 18, 2003)
416 ====================================================================
417
418 1. Fixed the INSTALL file to reflect the actual install paths.
419 Thanks to David Ronis <ronis@ronispc.chem.mcgill.ca> for
420 reporting the bug.
421
422 2. Fixed the configure script to only check for headers presence
423 instead of trying to compile them. This should fix issues
424 with old build environments. Thanks to Kari Lempiainen
425 <kari@funky.pp.fi> for reporting the bug.
426
427 3. Fixed restore to correctly ignore sockets when comparing
428 a dump (as socket cannot be properly restored anyway). Thanks
429 to Gunther Reiszig <gunther@mit.edu> for reporting the bug.
430
431 4. Fixed restore to correctly access the archive file (-A argument)
432 even when using a remote tape. Thanks to Fabrice Bellet
433 <fabrice@bellet.info> for reporting the bug.
434
435 5. Fixed (again) handling of long (largefile) seeks in rmt.
436 Thanks to Fabrice Bellet <fabrice@bellet.info> for reporting
437 the bug.
438
439 6. Fixed restore corner case when dealing with large block sizes
440 dump is able to write now (-b 1024). Thanks to Fabrice Bellet
441 <fabrice@bellet.info> for reporting the bug.
442
443 7. Fixed a bug preventing dump to access a filesystem having
444 a label exactly 16 bytes in length. Thanks to <minduni@ti-edu.ch>
445 for reporting the bug.
446
447 8. Made dump store dump dates using explicit timezones, fixing a
448 problem with incremental dumps if the timezone is modified
449 between the dumps. Thanks to Philippe Troin <phil@fifi.org> for
450 the bug report and the patch.
451
452 9. Fixed a bug encountered when dumping individual files (not full
453 filesystems or directories) and dangling symbolic links happen
454 to be in the list of files. For as far as dump is concerned,
455 dangling symbolic links are allowed, and are dumped as is.
456 Thanks to Jin-su Ahn <jsahn@ee.snu.ac.kr> for reporting the
457 bug and providing the fix.
458
459 10. Fixed open and creation modes and permissions for QFA and
460 table-of-contents files in dump and restore. Thanks to
461 Philippe Troin <phil@fifi.org> for the patch.
462
463 11. Fixed the archive file descriptor handling enabling it to be 0.
464 This can happen in some cases when shell redirections are used.
465 Thanks to Philippe Troin <phil@fifi.org> for the patch.
466
467 12. Delayed the opening of archive file until after suid had been
468 dropped (fixing a possible security issue if dump is suid).
469 Thanks to Philippe Troin <phil@fifi.org> for the patch.
470
471 13. Fixed the 'S' command handling in the rmt client part.
472 Thanks to Philippe Troin <phil@fifi.org> for the patch.
473
474 14. Modified the end-of-tape script handling to print out statistics
475 (and stop the timer) before launching the eot script. Also, the eot
476 script does not get run anymore when using -M (which makes sense) or
477 when multiple tapes are listed on the command line
478 (-f tape0,tape1,tapen) (which also makes sense).
479 Thanks to Philippe Troin <phil@fifi.org> for the patch.
480
481 15. Relicensed dump/restore under the 'revised' BSD license, as per
482 ftp://ftp.cs.berkeley.edu/ucb/4bsd/README.Impt.License.Change.
483
484 16. Added LZO compression to dump. This new compression method has
485 the advantage of being super fast, thus not killing tape streaming
486 on slow machines. Thanks to Helmut Jarausch
487 <jarausch@igpm.rwth-aachen.de> for the patch and to
488 Markus Oberhumer <markus@oberhumer.com> for giving special permission
489 to include his miniLZO project (GPL licensed) in dump/restore.
490
491 17. Some small buffer overruns fixes in rmt. Thanks to Antonomasia
492 <ant@notatla.demon.co.uk> for reporting the bugs.
493
494 18. Added a special rmt version which can do encryption when writing
495 to tape. Read examples/encrypted_rmt/README for details on
496 how to enable and configure it. Thanks to Ken Lalonde
497 <ken@globalremit.com> for the patch.
498
499 19. Made dump work with 2.5 kernel end of tape early warning semantics.
500 Thanks to Kai Makisara <Kai.Makisara@kolumbus.fi> for the patch.
501
502 20. Fixed a bug which caused dump -w|-W not to work anymore, because
503 the fs_freq and fs_passno fields in /etc/mtab are always set
504 to 0 0. Thanks to Trent Piepho <xyzzy@speakeasy.org> for
505 reporting the bug.
506
507 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
508 =======================================================================
509
510 1. Added a note in the restore man page clarifying the question
511 'set the permissions on the current directory ?' asked by
512 restore at the end of treatment in -i and -x modes.
513
514 2. Fixed the endianess issues when reading compressed tapes.
515 Thanks to Dark Force <daq4th@users.sourceforge.net> for
516 reporting this bug and providing test cases.
517
518 3. Fixed the "ACL won't be dumped" warning message (which showed
519 an extra, unrelated error message). Thanks to Dragan Krnic
520 <dkrnic@lycos.com> for reporting this bug.
521
522 4. Made dump look first into /etc/mtab, then into /etc/fstab
523 when searching for filesystem paths. Also fixed some problems
524 caused by binding mounts or multiple block device mounts.
525 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
526 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
527 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
528 and #160305, Sourceforge bugs #618699 and #679832).
529
530 5. Made dump's -I option accept the value '0' meaning all the
531 read errors will be ignored. This can be useful when running
532 dump from unattended sessions (like cron jobs). Thanks to
533 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
534
535 6. Fixed the output of dump to indicate 'blocks' instead of
536 'tape blocks' in the various messages (blocks are always
537 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
538 made it clearly print the current blocksize at the start of
539 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
540 the suggestions.
541
542 7. Made rmt understand long (largefiles) seeks.
543
544 8. Fixed build with very old versions of libext2fs, where
545 EXT2_FT_* constants were undefined.
546
547 9. Made dump accept the dumpdates path on the command line
548 (-D file option) instead of using only the hardcoded one.
549 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
550 suggestion.
551
552 10. Enabled rmt, LFS, readline, QFA options by default in
553 ./configure. Updated the configure process (new versions
554 of config.guess, config.sub etc).
555
556 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
557 =======================================================================
558
559 1. Changed dump to use fcntl(F_SETLK) style locking instead
560 of flock() when locking the dumpdates file. With the old
561 locking scheme, a local user having read rights on the
562 dumpdates file could be able to do a Denial of Service attack
563 on dump. In order to lock the dumpdates file with the new
564 scheme, the user would need to have write access on the file.
565 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
566 reporting the bug (originally a bugtraq post).
567
568 2. Fixed interactive 'ls' which caused spurious errors warnings
569 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
570 <jorgen_ostling@users.sourceforge.net> for reporting this
571 bug.
572
573 3. Fixed dump's estimate when dealing with sparse inodes.
574
575 4. Modified dump to allow setting a blocksize bigger than 32kB
576 (raised the limit to 1024kB), because newer hardware needs
577 this for proper operation. Thanks to Dirk Traenapp
578 <dtraenapp@users.sourceforge.net> for reporting this.
579
580 5. Fixed a bug causing Dump to stop and report an error if an
581 inode number in the exclude file was followed by some amount
582 of whitespace. Thanks to Jeffrey Sofferin
583 <sofferin@users.sourceforge.net> for reporting this bug.
584
585 6. Fixed a bug which caused restore, in some particular cases,
586 to ask some 'scary' questions and leave a bunch of RSTTMP
587 directories behind when restoring incremental tapes. Thanks
588 to Philippe Troin <phil@fifi.org> for reporting this bug and
589 providing the test cases.
590
591 7. Changed the wording when inodes are excluded from dump:
592 replaced 'Added inode 7 to exclude list' with
593 'Excluding inode 7 (resize inode) from dump', as suggested
594 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
595
596 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
597 ===================================================================
598
599 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
600 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
601 <eros@lamel.bo.cnr.it> for reporting the bug.
602
603 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
604 ===================================================================
605
606 1. Made dump print out the ext2 logical block number in case of a read
607 error, which can be given as an argument to debugfs' ncheck command,
608 in order to find out the affected inode. Added note about this
609 usage in dump's man page.
610
611 2. Fixed a problem in restore when reading tapes written on big
612 endian machines with very old versions of dump. The patch was
613 contributed by George Helffrich <george@geology.bristol.ac.uk>.
614
615 3. Fixed the tape length calculation when using large tapes
616 and compression. Thanks to Georg Lippold
617 <g_lippold@sourceforge.net> for reporting the bug.
618
619 4. Added a new set of examples in dump_on_cd_2 directory, based
620 on dump_on_cd examples but somewhat enhanced, supporting DVD
621 media, and localized in english and german. Thanks to
622 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
623
624 5. Made dump save 32 bit UID/GID. Internally, this was achieved
625 by switching from the old BSD inode format to the new BSD
626 inode format, which means that the tape format was changed.
627 However, since all restore versions out there should
628 transparently support both inode formats, the change should
629 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
630 reporting the bug.
631
632 6. Fixed a lot of warnings in the code shown when compiling
633 with 'gcc -W'. Thanks to Matthias Andree
634 <matthias.andree@stud.uni-dortmund.de> for reporting this.
635
636 7. Fixed a small markup bug in the dump man page. Thanks to
637 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
638 patch.
639
640 8. Rewrote entirely the man pages using the tmac.an macro
641 package (Linux man page format) instead of the original BSD
642 format. They should be now cleaner and easier to modify.
643
644 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
645 ==================================================================
646
647 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
648 non Linux clients were misinterpreted. The description of the problem
649 (incompatible numbering in Linux mtio opcodes) is documented at
650 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
651 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
652 providing an excellent, cross-platform replacement for rmt in his
653 star package.
654
655 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
656 which caused restore to fail when the inode maps were not entirely
657 contained on the first volume (when using really small volumes or
658 when dumping a huge number of inodes).
659
660 3. Fixed a problem in dump, where files in subdirectories of directories
661 that have the nodump flag set (or being excluded with -e/-E) were
662 sometimes incorrectly dumped (instead of being excluded). The original
663 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
664 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
665 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
666 help testing the patch.
667
668 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
669 which makes one able to pipe the output of dump, by the net, to
670 a remote CD-burner server.
671
672 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
673 and use GNU's symbolic syntax over rmt instead of numerical values
674 to assure multiple platform compatibility.
675
676 6. Documented the -d option in restore.
677
678 7. Added a -v (verbose) mode to dump. For now it just prints the number
679 of the inode being dumped, but this could evolve in future versions
680 to include interesting debugging output.
681
682 8. Added a -o flag to restore, which automatically restores the current
683 directory permissions in -i or -x mode, without asking the operator
684 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
685 and Peter Samuel <peters@e-smith.com>.
686
687 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
688 ====================================================================
689
690 1. Fixed a bug in the QFA feature which could have caused
691 a bad tape positionning on restore (causing extra delays in
692 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
693
694 2. Added a small note in the dump man page specifying that
695 there should be no white space between the option letter and
696 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
697
698 3. Made restore work with older versions of the readline library,
699 by checking in configure for several extended symbols. Restore
700 can now be compiled with a readline as old as the 2.0 release
701 (though it may be a good idea to upgrade it to a more recent
702 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
703 reporting the build failures.
704
705 4. Fixed a performance problem with the QFA file creation in
706 dump, which made unnecessary seeks on the tape slowing down
707 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
708 reporting this issue.
709
710 5. Removed the inclusion of some kernel headers in the dump
711 source, which prevented the compile in some kernel/glibc
712 headers/architecture combination. Thanks to Bdale Garbee
713 <bdale@gag.com> for reporting the bug.
714
715 6. Added the appropriate error message when dump fails to
716 open the output file for writing. Thanks to Amith Varghese
717 <amithv@yahoo.com> for reporting this bug.
718
719 7. Made restore able to understand large Solaris ufsdump tapes
720 (containing inodes bigger than 4194304). Sun have introduced
721 an "extension" to the dump tape format when dealing with
722 those inodes, which was uncorrectly handled by Linux restore.
723 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
724 providing a test case.
725
726 8. Added the -m parameter to dump which optimises the output for
727 inodes having been changed but not modified since the last dump
728 ('changed' and 'modified' have the meaning defined in stat(2)).
729 For those inodes, dump will save only the metadata, instead of
730 saving the entire inode contents. Inodes which are either
731 directories or have been modified since the last dump are saved
732 in a regular way. Uses of this flag must be consistent, meaning
733 that either every dump in an incremental dump set have the flag,
734 or no one has it.
735
736 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
737 =======================================================================
738
739 1. Fixed behaviour of dump when exceeding resource limits
740 (SIGXFSZ treatment).
741
742 2. Added the -L flag to restore to allow the user to specify a
743 maximal allowed number of miscompares when using restore
744 with the -C option to check the backup.
745
746 3. Detailed the manual entry for the -N option of restore.
747
748 4. Added the -a flag to restore to make able doing unattended
749 restores in -i or -x mode (automatically walks through the
750 multiple dump volumes).
751
752 5. Extended the QFA mode to work with local files and/or
753 remote tapes and files. This way, restore can know in advance
754 the tape number and the offset for the inodes to extract and
755 can minimize the extraction time by seeking directly to the
756 good tape/offset.
757
758 6. Added the -A <archive> option to both dump and restore,
759 which makes dump to archive a dump table-of-contents in
760 the specified file to be used by restore to determine
761 whether a file is in the dump file that is being restored.
762 (the archive file syntax is also compatible with the
763 Solaris ufsdump generated one).
764
765 7. Small fix in restore making it able to read some (broken ?)
766 Solaris ufsdump tapes.
767
768 8. Fixed dump to correctly recognise the root filesystem when using
769 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
770 for reporting this bug.
771
772 9. Added the -P <file> option to restore to create a
773 Quick File Access file from an already made dump. Patch
774 contributed by Uwe Gohlke <uwe@ugsoft.de>.
775
776 10. Made restore compile and run on Solaris, making it a
777 possible replacement for the standard ufsrestore. Port was
778 contributed by Uwe Gohlke <uwe@ugsoft.de>.
779
780 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
781 =====================================================================
782
783 1. Added a set of backup scripts from Eugenio Diaz
784 <getnito@yahoo.com> in the examples section. It features
785 automatic (cron based) full and incremental dumping of
786 several filesystems on a separate filesystem.
787
788 2. Fixed a off-by-one miscalculation which disabled dumping
789 a one letter subdirectory. Thanks to Chris Siebenmann
790 <cks@utcc.utoronto.ca> for reporting the bug.
791
792 3. Fixed several restore bugs occuring when trying to
793 restore some missing files on the tape. Thanks to Chris
794 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
795
796 4. Fixed --with-ldopts configure argument passing, installing from
797 a separate object directory, makefile cleanups contributed
798 by <splite@purdue.edu>.
799
800 5. Fix a bug which could caused, in some conditions, the highest
801 number inode of a filesystem, to not be dumped. Many thanks
802 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
803 chase this bug.
804
805 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
806 =======================================================================
807
808 1. Added a mini howto from Patrick Walsh in the examples
809 directory.
810
811 2. Minor man pages syntax corrections. Thanks to
812 Chris Danis <screechco@home.com> for reporting the bugs.
813
814 3. Added a script from David B. Peterson <dave@toppledwagon.com>
815 to the examples section. It features dumping several
816 filesystems to a remote tape drive upon ssh.
817
818 4. Added a patch provided by Richard Jones <rich@annexia.org>
819 which allows BRADEMAX (number of read errors tolerated by
820 dump) to be adjusted using the -I option.
821
822 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
823 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
824 patch.
825
826 6. Add the -l option to restore to specify if, when doing a
827 remote restore, the file used to access the backup is a
828 regular file (the defaults being a tape drive). Restore needs
829 to know this information when reading a remote compressed
830 dump. Previously, this information was autodetected, but
831 the autodetection code fails (with ioctl: Inappropriate ioctl
832 for device) when using a non Linux remote box. Thanks to
833 many users and especially to Eros Albertazzi
834 <eros@lamel.bo.cnr.it> for reporting this.
835
836 7. Found a workaroung for the dump deadlock problem (3 childs
837 stuck in pause(), father in read()). The workaround seems
838 to work for me and several beta-testers. If it doesn't work
839 for you, please report back.
840
841 8. Updated the RPM spec file (BuildPrereq, URL etc).
842
843 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
844 ========================================================================
845
846 1. Fixed the permissions of a newly created QFA file by dump.
847
848 2. Cleaned up the source of dump (the external variables
849 definition was a complete mess, making possible to have
850 objects overlap).
851
852 3. Fixed restore to use the full tape volume path when doing
853 a compare (since it changes the working directory to the
854 filesystem being compared in the process).
855
856 4. Added the -q option to dump which makes dump abort
857 whenever it needs operator attention. It should be
858 easier to use dump in scripts with this option.
859
860 5. Detect the use of incompatible options to dump and
861 refuse them (like -a and -B options together).
862
863 6. Added bzip2 compression to dump/restore (use option -j level
864 to select it). Note that you will need a restore version
865 >= 0.4b24 in order to restore a bzip2 compressed dump.
866 The same warning as for the zlib compression applies:
867 the tape format of a bzip2 dump is not compatible with the
868 original BSD tape format.
869
870 7. Fixed a overflow problem in dump corrupting the dump when
871 very large files were encountered. Thanks to Vladimir Nadvornik
872 for the bug report.
873
874 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
875 kernel buffer/page cache before starting the dump, helping
876 a bit those who use dump on mounted filesystems. Thanks to
877 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
878 for this suggestion.
879
880 9. Updated the RPM spec file following the RedHat changes
881 (dynamically linked binaries now in /usr/sbin etc).
882
883 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
884 which enables restore to recognise multi volume compressed dumps
885 done on CD. Included his scripts for dump (which pipe the dump
886 output directly into cdrecord) and restore. There is now
887 possible to to backups to CD on the fly!
888
889 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
890 ===================================================================
891
892 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
893 Marcus Meissner (Caldera International Security Dept.).
894
895 2. Implement the Sun rmt extended protocol. Patch submitted
896 by Ian Gordon <iangordon@users.sourceforge.net>.
897
898 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
899 the linux kernel header. This ensures that dump always has the
900 latest version of this file. Patch submitted by
901 Andreas Dilger <adilger@turbolinux.com>.
902
903 4. Report any filesystem present in either /etc/fstab with a
904 positive passno or /etc/dumpdates in dump -w output.
905 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
906
907 5. Fixed the looping problem in dump introduced in the
908 previous version.
909
910 6. Changed the -B option of dump to limit the size of
911 _compressed_ output per volume if compression is on.
912 Patch contributed by Helmut Jarausch
913 <jarausch@igpm.rwth-aachen.de>. Note however that, since
914 it is impossible to predict the size of the compressed
915 data before doing the compression, the -B limit is a bit
916 conservative.
917
918 7. Fixed a bug in reading the operator typed file/tape path for
919 the next volume in restore.
920
921 8. Implemented a "-F script" option to restore which permits the
922 user to specify a script which will be launched at the
923 beginning of each tape, useful for automatic programming of
924 tape changers for example. See the restore man page for the
925 script parameters and return codes.
926
927 9. Small fix for the QFA routines provided by Uwe Gohlke
928 <uwe@ugsoft.de>, and some recommendations for QFA uses in
929 the man pages.
930
931 10. Fixed the multivolume restoring where making a mistake
932 to the 'Mount next tape' prompt caused several blocks to
933 be lost.
934
935 11. Enhanced the -e option of dump to take as a parameter a
936 comma separated list of inode numbers.
937
938 12. Added the -E option to dump which specify a file containing
939 inode numbers to exclude from the dump.
940
941 13. Fixed the compressed multi-volume dump + restore.
942
943 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
944 ==================================================================
945
946 1. Made dump report the number of blocks dumped per volume.
947 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
948
949 2. Fix a bug in dump where waiting too much at the 'change volumes'
950 question caused the volume to be rewritten. Thanks to
951 Graham Reed <greed@users.sourceforge.net> for reporting the
952 bug and providing a patch.
953
954 3. Added a compression option to dump, contributed by Jerry
955 Peters <gapeters@worldnet.att.net>.
956
957 WARNING: the compression works only on files/pipes and
958 drives supporting variable block size.
959
960 WARNING2: tapes created using the compression feature are
961 incompatible with the standard dump tape format, and a
962 version of restore >= 0.4b22 will be needed for extraction.
963
964 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
965 architectures. Thanks to Paul Slootman <paul@debian.org> for
966 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
967 it upstream.
968
969 5. Many cleanups (CPP defines, const char warnings, check of
970 ext2fs COMPAT flags, time_t cleanups, added libext2 version
971 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
972
973 6. Made --prefix option work in configure. All the install path
974 are now based on the configure parameters.
975
976 7. Added the Quick File Access mode in dump/restore, contributed
977 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
978 a file tape position for each inode, and this file is used by
979 restore (if called with parameter Q and the filename)
980 to directly position the tape at the file restore is currently
981 working on. This saves hours when restoring single files from
982 large backups, saves the tapes and the drive's head. Use
983 --enable-qfa option of configure to compile in the QFA support.
984
985 8. Added the possibility to dump several files and directories
986 in a single invocation of dump. Thanks to Uwe Gohlke
987 <uwe@ugsoft.de> for implementing this option.
988
989 9. Fixed the dumping and restoring of files with holes
990 (files ending with holes didn't get dumped or restored
991 correctly).
992
993 10. Fixed a socket descriptor leak in dump, which leaved opened
994 3 file descriptors per dump process (and there is one dump
995 process per tape).
996
997 11. Fixed dump large file system support, by explicit use of
998 open64/lseek64/etc functions (explicit use needed because
999 e2fsprogs libraries don't behave well when compiled with
1000 FILE_OFFSET_BITS=64).
1001
1002 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
1003 ======================================================================
1004
1005 1. Fixed some bugs in the dump -w|-W logic introduced by
1006 the previous version. Thanks to Andreas Dilger
1007 <adilger@turbolinux.com> for his help on this one.
1008
1009 2. Fixed again a compilation problem when using old e2fs
1010 headers (filesystem label related). Thanks to many users
1011 who reported this stupid error.
1012
1013 3. Fixed a build problem on old lib5 based systems dealing with
1014 _PATH_MNTTAB being undefined. Thanks to John Adams
1015 <johna@onevista.com> for reporting the bug.
1016
1017 4. Improved the error detection in restore when creating
1018 the temporary files in TMPDIR. Now restore will corectly
1019 report a 'no space left on device' error instead of
1020 strange errors which could imply an archive corruption.
1021 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
1022 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
1023
1024 5. Added the throughput information in dump's progression
1025 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
1026 for the patch.
1027
1028 6. Use libext2fs's inode scanning routines, which are particularly
1029 robust when dealing with errors (especially when having some
1030 bad blocks on the filesystem). Many thanks to Colin
1031 <colin@horizon.com> for reporting the bug, investigating
1032 the issues, finding a workaround, writing the patch and
1033 fully testing it... (of course, if this patch will break
1034 anything, he is to blame for :-)).
1035
1036 7. Made dump and restore LFS aware. Dump can dump a filesystem
1037 containing large files, generate a large file on output and
1038 restore can restore them. This feature is not enabled by
1039 default, use --enable-largefile option of configure to enable
1040 it (you need to have a LFS aware glibc though). Thanks to
1041 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
1042 and to Theodore T'so <tytso@valinux.com> for his always
1043 useful thoughts.
1044
1045 8. Made dump ask upon a tape write error if it should rewrite
1046 the current volume (assume this is a bad tape) or if it should
1047 assume an end-of-tape condition (useful for tape drives which
1048 behaves badly at the end of the tape). Thanks to Andreas
1049 Dilger <adilger@turbolinux.com> for the suggestion.
1050
1051 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
1052 =======================================================================
1053
1054 1. Fixed a small compilation problem due to a change
1055 in the definintion of the struct sigaction in
1056 glibc 2.0 / libc5. Thanks to Gunther Schlegel
1057 <schlegel@riege.de> for reporting the bug and to
1058 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
1059 a fix.
1060
1061 2. Modified the label and uuid analysis in order to be
1062 self-contained (independant of kernel/libc headers). This
1063 should fix the compile with older kernel/libc headers and
1064 will preserve the functionality. Thanks to Bernhard Erdmann
1065 <bernhard.erdmann@gmx.de> for reporting the bug.
1066
1067 3. The 'exclude inode' option, if applied to a directory
1068 inode, excludes now correctly all the leaves of this
1069 directory. Thanks to John R. Dennison
1070 <gerdesas@users.sourceforge.net> for reporting the bug.
1071
1072 4. Fixed the '-e' option to disable the possibility
1073 to exclude the root inode (which causes the dumps to
1074 be unreadable by restore). Prevented array overflow
1075 when multiple -e options are used.
1076
1077 5. Fixed dump to correctly interpret a filesystem argument
1078 which is a mountpoint and it is not an absolute path
1079 (as specified in the fstab). Thanks to Bernhard R. Erdmann
1080 <be@berdmann.de> for reporting the bug.
1081
1082 6. Made dump able to backup files larger than 2 GB. Note that
1083 dump still doesn't cope with files larger than 4 GB.
1084
1085 7. Restore the real uid and gid before invoking an external
1086 RSH program (big hole when dump or restore is suid root!).
1087
1088 8. Get the values for the default tape device and the location
1089 of fstab file from the system headers. Thanks to
1090 Andreas Dilger <adilger@turbolinux.com> for the patch.
1091
1092 9. Made dump -w|-W report all recognized filesystems
1093 present in either /etc/fstab or /etc/dumpdates, and present
1094 the list in the same order as in fstab file. Thanks
1095 to Andreas Dilger <adilger@turbolinux.com> for the patch.
1096
1097 10. Made dump's -a (automatic end of tape detection) the
1098 default. Specifying one of -B, -c, -d or -s options will
1099 override the EOM detection. Thanks to Andreas Dilger
1100 <adilger@turbolinux.com> for the patch.
1101
1102 11. Save the ext2 filesystem label into the dump volume label.
1103 Specifying a volume label on the command line (-L option)
1104 will override this feature. Thanks to Andreas Dilger
1105 <adilger@turbolinux.com> for the patch.
1106
1107 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
1108 =====================================================================
1109
1110 1. Fixed the signal handling in dump (which I broke in 0.4b17)
1111 which was causing several strange dump failures (dump
1112 hanged or segmentation faults etc).
1113
1114 2. Specified the default blocksize in the dump man page.
1115
1116 3. Changed two info messages of restore to be written on stdout
1117 instead of stderr in order to leave stderr only for errors
1118 or warnings. Thanks to Stephen Atwell
1119 <satwell@urbana.css.mot.com> for the suggestion.
1120
1121 4. Corrected an off by one calculation which prevented
1122 dumping a directory having a 1 character name length.
1123 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
1124 for reporting the bug.
1125
1126 5. Reinforce dump in order to allow only level 0 and no
1127 -u option when dumping a subdirectory, and document
1128 this in the man page. Thanks to Bernhard Erdmann
1129 <bernhard.erdmann@gmx.de> for reporting the bug.
1130
1131 6. Fixed a small harmless bug in restore which caused
1132 in some conditions hard links to be restored several
1133 times (therefore generation some warning messages).
1134 Thanks to Kenneth Porter <shiva@well.com> for
1135 reporting the bug.
1136
1137 7. Updated the RPM spec file to the latest RedHat version,
1138 providing FHS packaging and other cosmetic changes.
1139 You will need RPM version >= 3.0.5 in order to build the RPM.
1140
1141 8. Updated the configure script to check for libtermcap
1142 before checking for libreadline, because we need this
1143 library in order to compile the readline support.
1144
1145 9. Made dump understand the LABEL= and UUID= notation
1146 both in /etc/fstab and on the command line. Note that
1147 you will need the /proc filesystem in order to use
1148 these notations. Thanks to Erik Troan <ewt@redhat.com>
1149 for providing the patch.
1150
1151 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
1152 ===================================================================
1153
1154 1. Fixed a potential buffer overflow in restore. Thanks
1155 to Stan Bubrouski <satan@fastdial.net> for reporting
1156 the bug.
1157
1158 2. Fixed a readline-related bug which prevented
1159 'cat DUMPFILE | restore -i -f -' from working. Thanks
1160 to Charles Karney <karney@users.sourceforge.net>
1161 for the bug report.
1162
1163 3. Changed a few "panic" into "exit", causing restore to
1164 be more stable against some attacks (like the last one
1165 reported on Bugtraq, although the last version of restore
1166 was not vulnerable - just dumped core). Thanks to
1167 Andreas Hasenack <andreas@conectiva.com.br> for reporting
1168 the bugs to me.
1169
1170 4. Removed the suid-root bit on dump and restore in the
1171 default build (and generated RPMs). It should be safer
1172 now. Users who need the suid-root capabilities in order
1173 to make network backups should read first the man page
1174 of dump and enable the suid bit manually.
1175
1176 5. Added -ltermcap to the compile parameters for restore
1177 when using readline support, in order to make the compile
1178 process work on some systems (SuSE ?). Thanks to
1179 Patrik Schindler <poc@pocnet.net> for reporting the bug.
1180
1181 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
1182 ====================================================================
1183
1184 1. The -F script is called now *only* at the end of a tape,
1185 not at the end of the dump. Thanks to Christian Haul
1186 <haul@informatik.tu-darmstadt.de> for the bug report.
1187
1188 Normally, the device name and the current volume number
1189 are passed on the command line of the script. However,
1190 if you want the old style script (with no arguments
1191 passed on the command line), you can enable it in
1192 configure with the --enable-oldstylefscript.
1193
1194 2. Use posix signal handling to preserve dump functionality
1195 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
1196 the patch.
1197
1198 3. Made the exit code of restore in case of a 'C'ompare
1199 command reflect the result of the compare. An exit status
1200 of 0 means the dump archive is correct, 1 means tape errors,
1201 2 means that some files were modified. Thanks to Kenneth Porter
1202 <shiva@well.com> for the suggestion.
1203
1204 4. Made (finally) quotes work in interactive restore.
1205
1206 5. Small fixes in order to allow dump to compile with a
1207 really old e2fsprogs version. Thanks to Ian Zimmerman
1208 <itz@speakeasy.org> for the bug report.
1209
1210 6. Add GNU readline capabilities to interactive restore.
1211 Use configure's --enable-readline flag to enable this feature.
1212 Thanks to Patrik Schindler <poc@pocnet.net> for the
1213 suggestion.
1214
1215 7. Do the compare on the fly in restore 'C' mode (this will
1216 allow not to exhaust the available /tmp space when
1217 ccmparing large files). Thanks to Kenneth Porter
1218 <shiva@well.com> for the suggestion.
1219
1220 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
1221 ====================================================================
1222
1223 1. Fixed some several duplicate 'DUMP: DUMP:' in the
1224 output of dump.
1225
1226 2. Corrected the estimation of blocks to dump. Note that
1227 this is still wrong for large directory inodes, due
1228 to the size of a BSD directory entry on the tape
1229 (estimation is based on the size of the ext2 directory,
1230 which is smaller because it doesn't contain the null
1231 character at the end).
1232
1233 3. Made dump report the total number of MB written to
1234 tape at the end of run. Thanks to W. Reilly Cooley
1235 <wcooley@nakedape.cc> for the patch.
1236
1237 4. Added the -X option to restore giving the possibility
1238 to read the names of the files to be extracted/listed
1239 from a text file (in addition of the command line).
1240 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
1241 patch.
1242
1243 5. Added the device name and the current volume number
1244 as arguments to the end of tape script (dump -F option).
1245
1246 6. Made the multi-volume dump work again (after having
1247 broken it in 0.4b15).
1248
1249 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
1250 ===================================================================
1251
1252 1. Added a prompt command in interactive restore mode. Thanks
1253 to Andreas Dilger <adilger@home.com> for the patch.
1254
1255 2. Fixed a buffer overflow problem in dump (caused by
1256 not checking the size of the filesystem parameter).
1257 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
1258 reporting this on Bugtraq (and to several dump users
1259 who forwarded me his mail).
1260
1261 3. Added the '-F script' option to dump in order to
1262 launch a script at the end of each tape (to be used
1263 with a tape changer, or to notify the sysadmin by
1264 pager etc.).
1265
1266 4. Fixed a bug in restore compare code caused by the changes
1267 I made in 0.4b14.
1268
1269 5. Fixed the treatment of options using the old BSD syntax
1270 in both dump and restore.
1271
1272 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
1273 =======================================================================
1274
1275 1. Fixed a bug in dump which may have caused invalid deleted
1276 directories to be dumped out if they were deleted after the
1277 mapping pass. This could occure on active filesystem and lead
1278 to heap corruption (causing dump malfunction in many possible ways).
1279 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
1280 bug and submitting the patch.
1281
1282 2. Fixed the handling of the filesystem parameter in dump. A
1283 '/mnt/disk' parameter caused the disk contents to be dumped,
1284 but a '/mnt/disk/' parameter caused the mountpoint directory
1285 to be dumped (generally an empty directory).
1286
1287 3. Improved the output of dump in order to tell which directory
1288 it is currently dumping (when dumping a subtree).
1289
1290 4. Added the '-e' exclude inode option to dump. Thanks to
1291 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
1292
1293 5. Added a REPORTING-BUGS file in order to provide a guide
1294 on how to correctly report a bug in dump/restore.
1295
1296 6. Fixed a restore problem when restoring a hard link to an inode
1297 having the immutable or append-only attribute set. Thanks to
1298 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
1299
1300 7. Fixed a compatibility problem between dumps made on little
1301 endian machines (the format was correct) and big endian
1302 machines (the format was incorrect). This fix break the
1303 compatibility with the older dumps made on big endian
1304 machines (sparc, m86k, ppc etc). For the first time in
1305 linux dump's history (I believe), the dumps made by *BSD,
1306 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
1307 so interchangeable. Thanks to Rob Cermak
1308 <cermak@ahab.rutgers.edu> for submitting the bug and
1309 helping me test the fix.
1310
1311 8. Fixed the way dump reports the remaining percent/time, if
1312 the number of blocks actually dumped exceeds the estimated
1313 number of blocks. Thanks to Jean-Paul van der Jagt
1314 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
1315
1316 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
1317 ======================================================================
1318
1319 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
1320 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
1321
1322 2. Removed the requirement to build the RPM as root from the
1323 spec file. Thanks to Christian Weisgerber
1324 <naddy@mips.rhein-neckar.de> for submitting this.
1325
1326 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
1327 causing dump to block when dumping really huge filesystems.
1328 Many thanks to Patrik Schindler <poc@pocnet.net> for
1329 helping me find this bug.
1330
1331 4. Fixed the treatment for an interrupt signal when dump access
1332 the remote tape through RSH. Thanks to Christian Weisgerber
1333 <naddy@mips.rhein-neckar.de> for providing the patch.
1334
1335 5. Fixed a bug which was causing dump/restore to display
1336 garbage characters instead of the remote host name.
1337
1338 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
1339 =====================================================================
1340
1341 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
1342 <kukuk@suse.de> for submitting the patch.
1343
1344 2. Fix for the exit code when using the size estimate option of
1345 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
1346 submitting the patch.
1347
1348 3. Handle EINTR in atomical reads/writes in dump, which was causing
1349 dump to fail on some systems. Thanks to Eric Jergensen
1350 <eric@dvns.com> for reporting the bug and submitting the patch.
1351
1352 4. Handle more than 16 characters for the device names in dumpdates.
1353 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
1354 tracking down the problem and proposing the solution.
1355
1356 5. Fixed a bug in dump which prevented the creation of the
1357 dumpdates file when doing a 0-level dump without already
1358 having a dumpdates file. Thanks to Patrik Schindler
1359 <poc@pocnet.net> for reporting the bug.
1360
1361 6. Changed the way dump 'S' flag reports the size estimate
1362 from number of blocks into bytes (making it compatible
1363 with the Solaris version, and simplifying things for
1364 amanda users). Thanks to Jason L Tibbitts III
1365 <tibbs@math.uh.edu> for reporting the bug.
1366
1367 7. Fixed a compatibility problem in linux/alpha dump tape format.
1368 Now the linux/alpha dump are (again) compatible with the
1369 other dump formats. But this breaks compatibility with
1370 older dumps made on alpha. Thanks to Mike Tibor
1371 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
1372
1373 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
1374 ======================================================================
1375
1376 1. Added a '--enable-kerberos' to configure.
1377
1378 2. Added a 'S' option to dump which determines the amount of space
1379 that is needed to perform the dump without actually doing it, similar
1380 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
1381 <cermak@ahab.rutgers.edu>.
1382
1383 3. Added a 'M' multi-volume option to dump and restore which enables
1384 dumping to multiple files (useful when dumping to an ext2
1385 partition to make several dump files in order to bypass the 2GB
1386 file size limitation). The 'f' argument is treated as a prefix and
1387 the output files will be named <prefix>001, <prefix>002 etc. With
1388 the 'M' flag, restore automatically selects the right file without
1389 asking to enter a new tape each time.
1390
1391 4. Fixed a memory leak which was causing dump to grow very big
1392 (270MB when dumping a 10GB filesystem...). Thanks to Jason
1393 Fearon <jasonf@netrider.org.au> for reporting the bug.
1394
1395 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
1396 ======================================================================
1397
1398 1. Make configure test if the system glob routines support
1399 extended syntax (ALTDIRFUNC). If not, use the internal glob
1400 routines instead of system ones. Thanks to Bernhard Sadlowski
1401 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
1402 and helping me resolve this and other minor libc5 compiling
1403 glitches.
1404
1405 2. Fix a problem when dumping a ext2fs with the 'filetype'
1406 feature enabled. Thanks to Patrick J. LoPresti
1407 <patl@cag.lcs.mit.edu> for reporting the bug and to
1408 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
1409
1410 3. Made the nodump flag work on directories. A directory which
1411 has the nodump flag gets never dumped, regardless of its
1412 contents.
1413
1414 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
1415 which allows dump on an active ext3 filesystem. However, this
1416 is a "quick and dirty" patch which enables backup of an ext3
1417 filesystem through the ext2 compatibility (by ignoring the
1418 NEEDS_RECOVERY bit). The journal file is not recognized and
1419 it is dumped (it should not).
1420
1421 5. Test the superblock compatibility flags when dumping, in order
1422 to be sure that we know how to deal with specific features.
1423
1424 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
1425 ====================================================================
1426
1427 1. Use lchown instead of chown, fixing a possible security problem
1428 when restoring symlinks (a malicious user could use this
1429 to deliberately corrupt the ownership of important system files).
1430 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
1431 this and providing the patch.
1432
1433 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
1434 ====================================================================
1435
1436 1. Put dump sources under CVS, added Id tags in all files so
1437 one can use 'ident' on binary files.
1438
1439 2. Added the dump/restore version in the usage text so one can
1440 easily verify the version he is using.
1441
1442 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
1443 a va_start/va_end problem on linux-ppc (always call va_start
1444 va_end in pairs each time we use a vararg function).
1445
1446 4. Added again the DT_* constants because old libc does not
1447 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
1448 for submitting the bug report.
1449
1450 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
1451 this should enable dumping big (huge) filesystems.
1452
1453 6. Added the RSH environment variable in order to be able to
1454 use a rsh replacement like ssh when doing remote backups (and
1455 bypass the security limitations of rcmd). Now you can do remote
1456 backups without being root (or making dump setuid root).
1457
1458 7. Modified again the way dumpdates works. For incremental dumps,
1459 we need to read dumpdates even if we are not using 'u' option.
1460 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
1461 this should work.
1462
1463 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
1464 ===================================================================
1465
1466 1. Removed the 'k' flag from the restore 'about' text if kerberos
1467 was not compiled in.
1468
1469 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
1470 to them (fsetflags takes a char*, setflags an open fd!).
1471
1472 3. (f)setflags is called only if the flags aren't empty. If the
1473 file is a special file, a warning is printed, because changing
1474 flags implies opening the device. Normally, a special file
1475 should not have any flag... (Debian bug #29775, patch provided
1476 by Abhijit Dasgupta <abhijit@ans.net>).
1477
1478 4. Made possible to dump a file system not mentioned in /etc/fstab.
1479 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
1480
1481 5. Changed the default behaviour to not create dumpdates
1482 unless 'u' option is specified. Removed the old "debian-patch"
1483 which provided the same thing. (Debian bug #38136, #33818).
1484
1485 6. Removed all those dump*announce, since they were getting old...
1486
1487 7. Added warning messages if dumpdates does not exist and
1488 when an inferior level dump does not exist (except for a level 0
1489 dump).
1490
1491 8. Debugged the glob calls in interactive mode: restore used a
1492 dirent struct which was different from the /usr/include/dirent.h
1493 one (this used to work, is it a glibc2 change?), so none of the
1494 compat glob (which used /usr/include/dirent.h) or the system glob
1495 worked. Restore use now the system dirent (and the system
1496 DT_* constants), which are compatible with BSD ones.
1497
1498 9. Added a configure flag (--with-dumpdatespath) to specify
1499 the location of dumpdates file. By default, it is
1500 /etc/dumpdates.
1501
1502 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1503 included the current date/version in man pages.
1504
1505 11. Corrected the estimation of remaining time when
1506 the operator doesn't change the tapes quickly enough. This
1507 was an old bug, I thought I corrected it, and discovered
1508 that in fact it was corrected in two different places, so
1509 the results canceled each other...
1510
1511 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1512 ===================================================================
1513
1514 1. Integrated multiple patches from RedHat, Debian and SuSE:
1515
1516 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1517 isn't specified.
1518 - several fixes in the man pages.
1519 - update the default tape device to /dev/st0.
1520 - many updates for Linux Alpha (byte ordering, size_t etc).
1521 - buffer overruns.
1522 - use environment variable for TMPDIR (instead of /tmp).
1523 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1524 - workaround egcs bug (RedHat bugs #4281 and #2989).
1525 - wire $(OPT) throughout Makefile's.
1526
1527 2. Upgrade the dump revision to 1, making possible to dump filesystems
1528 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1529
1530 3. Fix some compile warnings, prototype all functions.
1531
1532 4. Use glibc err/glob instead of internal compatibility
1533 routines (only if available).
1534
1535 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1536 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1537
1538 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1539 ======================================================================
1540
1541 1. Integrated the changes from FreeBSD-3.1-RELEASE
1542 (mostly bug fixes, buffer overruns, dump has now an "automatic
1543 tape length calculation" flag, dump/restore can use kerberos now
1544 (this is NOT tested), use environment variables for TAPE and
1545 RMT etc.).
1546
1547 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1548
1549 3. Corrected a bug in restore when using 'C' option with multi-volumes
1550 tapes (files splited accros two tapes give "size changed" errors
1551 when comparing).
1552
1553 4. Corrected the long standing bug when dumping multiple tapes.
1554 This works for me, needs further testing.
1555
1556 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1557 ====================================================================
1558
1559 1. Dump now runs correctly on kernels 2.1.x
1560 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1561
1562 Changes between versions 0.4b2 and 0.4b3
1563 ========================================
1564
1565 1. Use realpath() if available
1566
1567 2. Report statistics
1568
1569 Changes between versions 0.4b1 and 0.4b2
1570 ========================================
1571
1572 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1573 the patch)
1574
1575 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1576
1577 3. Fixed configure.in to correctly handle --enable-rmt
1578
1579 Changes between versions 0.3 and 0.4b1
1580 ======================================
1581
1582 1. Integrated the changes from 4.4BSD-Lite2
1583
1584 2. Integrated the patches from Debian and Red Hat
1585
1586 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1587
1588 4. Changed dump to use the Ext2fs library to get block addresses. This
1589 should solve the endianness problem on SparcLinux.
1590
1591 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1592 distribution's one) to use autoconf
1593
1594 6. Fixed a few minor bugs
1595
1596 Changes between versions 0.2e and 0.2f
1597 ======================================
1598
1599 1. Added the creation of named pipes (fifos) in restore.
1600
1601 2. Added the -N flag in the restore manual page.
1602
1603 3. Added the file kernel.patch which contains the llseek() optimization
1604 patch for 1.2.x kernels.
1605
1606 4. Fixed a bug in the restoration of symbolic links: owner and group were
1607 not restored.
1608
1609 5. Integrated some changes from FreeBSD 2.2.
1610
1611 6. Added a call to ftruncate() after restoring each file to restore
1612 correctly files ending by a hole.
1613
1614 Changes between versions 0.2d and 0.2e
1615 ======================================
1616
1617 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1618 with owner = root (0) and group = root/wheel/whatever (0).
1619
1620 Changes between versions 0.2c and 0.2d
1621 ======================================
1622
1623 1. Dump is now able to backup 2GB+ filesystems.
1624
1625 2. Dump and restore can now be linked as static binaries.
1626
1627 Changes between versions 0.2b and 0.2c
1628 ======================================
1629
1630 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1631
1632 Changes between versions 0.2a and 0.2b
1633 ======================================
1634
1635 1. Really fixed the bug that I should have corrected in 0.2a.
1636
1637 2. Enabled optimization again.
1638
1639 Changes between versions 0.2 and 0.2a
1640 =====================================
1641
1642 1. Disabled the optimization during compilation.
1643
1644 Changes between versions 0.1 and 0.2
1645 ====================================
1646
1647 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1648 the fs_type field (actually, I modified the file fstab.c to make it
1649 use the mntent functions).
1650
1651 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1652 made by dump should be readable by the BSD restore and backups made
1653 by the BSD dump should be readable by restore. Unfortunately, this
1654 means that the dump format has changed between version 0.1 and version
1655 0.2 :-(
1656
1657 3. Dump is now able to backup a subtree, it is no longer limited to whole
1658 filesystems like the BSD version.
1659
1660 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1661 than 2 GB.
1662
1663 Changes between versions 0.0 and 0.1
1664 ====================================
1665
1666 1. Now create links rdump and rrestore during the `make install' step.
1667
1668 2. Linux port specific bugs added to the manual pages
1669
1670 3. Incorrect estimation of the number of tapes blocks fixed when doing
1671 incremental backups.
1672
1673 4. Better ls-like format in restore in interactive mode.