Prepare for release 0.4b43
[dump.git] / CHANGES
1 $Id: CHANGES,v 1.317 2010/06/11 11:29:27 stelian Exp $
2
3 Changes between versions 0.4b42 and 0.4b43 (released June 11, 2010)
4 ===================================================================
5
6 1. Fix a bug in dump making impossible to handle large toc files
7 (> 2 GB). Thanks to X DUGi <xdugi@users.sourceforge.net> for
8 reporting the bug (Sourceforge bug #2820629)
9
10 2. Fix 'restore -x' in multi-volume mode, which caused files being
11 spanned on the first and second volumes to be incorrectly extracted.
12 Thanks to Adam Tkac <atkac@redhat.com> for reporting the bug.
13 (there may be several bug reports about this issue, including
14 Sourceforge bug #2117008, RedHat bugs #232415 and #444958).
15
16 3. Remove -lselinux -lsepol from the standard libraries list,
17 they shouldn't be needed if selinux is not enabled in dump
18 (this also fixes a build failure when selinux libraries are
19 not installed).
20
21 4. Fix restoration of extended attributes of fifos and device nodes.
22 Thanks to Igor Zhbanov <izh1979@gmail.com> for the bug report.
23
24 5. Don't attempt to set extended attributes on files that were not
25 extracted in 'restore -N' mode. Thanks to Jan Görig
26 <jgorig@users.sourceforge.net> for the patch.
27
28 6. Make the dump man page explicit about ext4 support.
29
30 7. Fix a bug in dump which caused EA entries to be unrecognizable by
31 restore in some corner cases. Add a workaround to restore making
32 it able to properly deal with those corrupted EA entries. Thanks
33 to John Austin <jaustin1@users.sourceforge.net> for reporting
34 the bug.
35
36 8. Fix some issues when restoring a dump which was generated using
37 exclusion patterns (either via -e or via the nodump attribute)
38 (see the Debian bug #574667 for details. Thanks to Frédéric
39 Brière <fbriere@fbriere.net> for the bug report and the associated
40 patch.
41
42 9. Fix progress/estimated blocks display in dump when doing really
43 huge backups. Thanks to Steve Bonds <sbonds@users.sourceforge.net>
44 for the bug report and tests. (Sourceforge bug #2987758)
45
46 10. Improve level 1 dump speed by rearanging the "mapdirs" code.
47 Thanks to Andreas Kies <andikies@t-online.de> for finding the
48 bottleneck and sending a patch (Sourceforge bug #2998119).
49
50 11. Use only ctime in order to compare newness instead of both
51 ctime and mtime. This should both speed up dumping time and
52 correct some side-effect bugs (like dumping files with an mtime
53 in the future in every incremental backup). Thanks to
54 Kieran Clancy <codebeard@users.sourceforge.net> for reporting the
55 bug (Sourceforge bug #2999207).
56
57 12. Extract dumped UNIX sockets instead of ignoring them.
58 (Sourceforge bug #3007216).
59
60 13. Compiler warning fixes (mainly warn_unused_result ones).
61
62 Changes between versions 0.4b41 and 0.4b42 (released June 18, 2009)
63 ===================================================================
64
65 1. Fix printout of the 'Connection to' message in dump/restore.
66 Thanks to Dale Wiles <dwiles@users.sourceforge.net> for the
67 bug report and its fix.
68
69 2. Fix dump -w/-W output for filesystems having the last backup
70 level equal to 0. Thanks to Pascal Bouchareine
71 <kalou@users.sourceforge.net> for reporting the bug.
72
73 3. Enable restore to handle restoring onto a different SELinux
74 policy from the dump. Thanks to Tony Nelson
75 <tony_nelson@users.sourceforge.net> for the patch.
76 (see http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=189845
77 for details)
78
79 4. Fix EA set failures when restoring immutable files. Thanks
80 to Andrew Kroeger for the patch.
81
82 5. Fix "mode file too large" errors when restoring huge backups.
83 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
84 report.
85
86 6. Add the ability to link against libtinfo library if present,
87 instead of libtermcap or libncurses libraries. Thanks to
88 Adam Tkac <atkac@redhat.com> for the patch.
89
90 7. When comparing a backup, do not consider to be an error the fact
91 that we encounter files created while dump was in progress.
92 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
93 report.
94
95 8. Force '-a' to be the default value when running 'restore -C'.
96 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
97 report.
98
99 9. Many improvements to the 'cron_dump_to_disk' example.
100 Thanks to Aaron S. Hawley <ashawley@users.sourceforge.net> for
101 the patch.
102
103 10. Fix the QFA generation when extended attributes are backuped
104 along with the inodes. Prior to this fix, the entries for
105 some inodes may get corrupted (duplicated entries or incorrect
106 ones), making the QFA file unusable for repositionning in
107 restore. Many thanks to Kenneth Porter <shiva@sewingwitch.com>
108 for the bug report and the testing of the fix.
109
110 11. Fix EA restoration in debug mode (Sourceforge bug #1986216).
111
112 12. Stefan Auracher <star7@onlinehome.de> noticed that the Tower
113 of Hanoi backup strategy documented in the man page wasn't
114 actually based on the Tower of Hanoi algorithm. The man page
115 was updated accordingly.
116
117 13. Fix missing level information from dump output when using
118 the default level (RedHat bug #493635). Thanks to Adam Tkac
119 <atkac@redhat.com> for the patch.
120
121 14. Fix the dump man page to reflect the fact that the default
122 dump level is 0 not 9 (RedHat bug #356121). Thanks to Adam Tkac
123 <atkac@redhat.com> for the patch.
124
125 15. Fixed a few spelling errors in the man pages (RedHat bug #489853).
126 Thanks to Adam Tkac <atkac@redhat.com> for the patch.
127
128 16. Fixed restore man page and program usage which incorrectly stated
129 that -P and -A were both allowed in the same invocation (RedHat
130 bug #490627). Thanks to Adam Tkac <atkac@redhat.com> for the patch.
131
132 17. Use sys_clone under Linux to share I/O contexts between dump
133 processes, thus drastically increasing the performance of dump
134 under CFQ (which is the default I/O scheduler used in a number of
135 distributions). Thanks to Jeff Moyer <jmoyer@redhat.com> for the
136 patch.
137
138 18. Add (preliminary) ext4 support - thanks to libext2fs which does
139 all the job for us. Thanks to Gertjan van Wingerde
140 <gwingerde@gmail.com> for the patch.
141
142 Changes between versions 0.4b40 and 0.4b41 (released January 2, 2006)
143 =====================================================================
144
145 1. Fix restore of dumped Access Control Lists. The previous
146 code for EA works fine, but ACLs needed conversion from the
147 ext2/3 disk format to posix_acl format before restoring.
148
149 2. Fix some issues with restoration of EA on big endian
150 platforms.
151
152 3. Fix restore when the symtab is over 2GB in size.
153
154 4. Made the directory hash indexing an optional feature,
155 accessible by the '-H' option of restore, and disable it
156 by default.
157
158 5. Fixed dump to not include extended attributes information
159 in the toc (archive) file which confused restore -t.
160
161 Changes between versions 0.4b39 and 0.4b40 (released May 2, 2005)
162 =================================================================
163
164 1. Changed restore to emit warnings (instead of emitting a fatal
165 error) if a file (or a directory) is unavailable for a
166 comparision (if the user doesn't have the necessary permissions
167 to access it for example). Thanks to Kenneth Porter
168 <shiva@sewingwitch.com> for the bug report.
169
170 2. Re-done the 'do not save directory entries to non-dumped inodes
171 (excluded from dump)' feature. The previous implementation
172 worked well for excluded directories but not for regular files.
173 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
174 report.
175
176 3. Fixed a bug in dump where the tape size was miscalculated when
177 the user used -d/-s to specify the tape characteristics. Thanks
178 to Philip Goisman <goisman@physics.arizona.edu> for reporting
179 the bug.
180
181 4. Fixed another bug introduced in restore with the hashtree
182 implementation. This one caused restore to stop saying
183 "removenode: non-empty directory" in some cases.
184
185 5. Added support for dumping and restoring ext2/3 extended
186 attributes (EA), like the access control lists (ACL) or
187 the security labels used by SELinux.
188
189 Changes between versions 0.4b38 and 0.4b39 (released January 21, 2005)
190 ======================================================================
191
192 1. The newly added dump_on_cd_3 example was buggy, replace it
193 with an updated version from Andrew Basterfield
194 <bob@cemetery.homeunix.org>.
195
196 2. Made restore to chdir() back into the initial directory when
197 dumping core while aborting a comparision operation. The
198 previous behaviour was to write the corefile at the root of
199 the directory being compared, which could very well be
200 read only and preventing the corefile generation. Thanks
201 to Kenneth Porter <shiva@sewingwitch.com> for the bug report.
202
203 3. Silenced the failure to call fgetflags() when comparing an
204 entry which has no ext2 attributes (as in lsattr()).
205
206 4. Fix a brown paper bug in restore -C which broke restore and
207 caused modifications on the filesystem being compared
208 (directories containing a file with the same name as the
209 directory get renamed to RSTTMP...). Thanks to Kenneth Porter
210 <shiva@sewingwitch.com> for finding the bug and helping me
211 reproduce it.
212
213 5. Made restore -C force the -N flag (no writing allowed on
214 the disk) in order to prevent more bugs like the above one.
215
216 Changes between versions 0.4b37 and 0.4b38 (released January 7, 2005)
217 =====================================================================
218
219 1. Fix a couple of troff syntax bugs in the man pages.
220 Thanks to Eric Raymond <esr@thyrsus.com> for the patch.
221
222 2. Made restore use either libncurses or libtermcap, depending
223 on which one is available at configure time.
224
225 3. Fixed restore negative size display bug when comparing a
226 dump containing files over 2GB. Thanks to Steve Bonds
227 <sbonds@users.sourceforge.net> for the bug report.
228
229 4. Do not save directory entries to non-dumped inodes
230 (excluded from dump). This will eliminate the 'missing
231 file' warnings when doing 'restore -C'.
232
233 5. Fix dump crash when backuping a huge (2TB) filesystem,
234 due to a bogus calculation on the inode map size.
235 Thanks to Kevin B. Haines <K.B.Haines@rl.ac.uk> for
236 submitting the bug and testing the fix.
237
238 6. Fix a problem in restore where the final \0 in the symbolic
239 link names could have been lost, generating corrupt filenames.
240 Thanks to Kyle Wilson <kyle.wilson@amd.com> for reporting the
241 bug.
242
243 7. Implemented a hash list for the directory names in restore.
244 The linear list used before caused problems in interactive
245 restores when dealing with directories having thousands of
246 entries. Thanks to Brian Ristuccia <bristuccia@starentnetworks.com>
247 for reporting the bug.
248
249 8. Improved restore -C, this time including the directory
250 attributes into the comparision.
251
252 9. Made restore understand tapes containing EA/ACLs (which will
253 be dumped by the next version of dump). In this version
254 extended attributes on the tape are ignored, for full EA/ACL
255 support wait for the next version or try the experimental EA
256 patch.
257
258 Changes between versions 0.4b36 and 0.4b37 (released July 7, 2004)
259 ==================================================================
260
261 1. Added the --enable-staticz configure option which enables
262 dump and restore to be linked with static versions of
263 libz and libbz2 (and dynamic versions of all other libraries).
264 This will make Debian users happy, because libz and libbz2
265 were the only needed libraries living in /usr, all the
266 others live in /lib. In case of system emergency, it is
267 better not to have to rely on an extra filesystem.
268 Thanks to Bdale Garbee <bdale@gag.com> for the suggestion.
269
270 2. Fix compilation on (at least the Linux Debian port to) AMD64.
271 (<ext2fs/ext2_types.h> defines some types (__s64 and __u64)
272 that are also defined by <linux/types.h> (<asm/types.h>) and
273 they conflict).
274
275 3. Make dump's reading of the dumpdates file a bit more robust,
276 preventing dump from crashing when the dumpdates file has
277 been modified by hand.
278
279 4. Fixed some offset calculations in dump code which could
280 lead to "bread lseek errors" on large filesystems. Thanks
281 to Bruce Lowekamp <lowekamp@users.sourceforge.net> for
282 reporting this bug and debugging the issue.
283
284 5. Made dump use the blkid library when searching for devices
285 by label or uuid instead of dump's own routines.
286
287 6. Corrected a bug in dump where a wrong LABEL=... line in
288 /etc/fstab could prevent dump from dumping unrelated
289 filesystems. Thanks to Bruce Lowekamp
290 <lowekamp@users.sourceforge.net> for reporting the bug.
291
292 Changes between versions 0.4b35 and 0.4b36 (released April 21, 2004)
293 ====================================================================
294
295 1. Fixed dump compilation with old gcc versions. Thanks to
296 Mike Castle <dalgoda@users.sourceforge.net> for the patch.
297
298 2. Fixed some warnings (howmany, roundup, powerof2 redefined)
299 when compiling against a recent glibc version.
300
301 3. Fixed a bug in restore preventing the read of a dump tape
302 written with Solaris 7 ufsdump. Thanks to Patrick Higgins
303 <phiggins@transzap.com> for reporting the bug and providing
304 the test case.
305
306 4. Changed dump to enable the creation of volumes of different
307 sizes in a single run (make -B accept a list of values).
308 Patch contributed by Florian Zumbiehl <florz@gmx.de>.
309
310 5. Use the glibc provided minor() and major() macros instead
311 of our own bitmask implementation. This should be safe for
312 when the major/minor namespace will migrate to 32 bits.
313 Thanks to Zhang Jun <zhangjun@nanjing-fnst.com> for reporting
314 the bug.
315
316 6. Made explicit in the dump man page that dump will not create
317 a remote file, it will only write to an already existing one.
318
319 7. Another try at making size estimates better again.
320
321 8. Put back the inconditional running of the end-of-tape
322 script which was changed in 0.434 to be run only when -M
323 or multiple -f were NOT specified. Some users rely on this
324 feature even when it is combined with -M/-f.
325
326 9. Fixed restore when restoring huge backups (where rstdir...
327 temporary files are over 2GB). Thanks to
328 Raphael Jaffey <rjaffey@artic.edu> for reporting this,
329 debugging the issue and making the patch.
330
331 10. Made restore understand FreeBSD UFS2 tapes. Thanks to
332 David <vrtin@users.sourceforge.net> for submitting the bug
333 and providing a test case.
334
335 11. Made dump work with an arbitrary integer level (no more 10
336 levels only limitation). Thanks to Matthew
337 <msvincen@midway.uchicago.edu> for the patch.
338
339 Changes between versions 0.4b34 and 0.4b35 (released December 21, 2003)
340 =======================================================================
341
342 1. Added a note in the dump man page saying that the default
343 blocksize can be 32 if -d is used with a high density argument.
344 Thanks to Antonios Christofides <A.Christofides@itia.ntua.gr>
345 for the patch.
346
347 2. Fixed configure to correctly understand CPPFLAGS, CFLAGS,
348 LDFLAGS environment variables. Thanks to Arcady Genkin
349 <antipode@users.sourceforge.net> for reporting the bug.
350
351 3. Made -e/-E options of dump accept an unlimited number of inodes
352 to be excluded, rather than a hardcoded maximum. Thanks to
353 Dietrich Rothe <d-rothe@users.sourceforge.net> for the patch.
354
355 4. Updated the autoconf system to 2.50. Forced the -D_BSD_SOURCE
356 and -D_USE_BSD_SIGNAL defines in configure in order to solve
357 64bit build problems because quad_t is redifined with a
358 different signature. Thanks to Mike Harris <mharris@redhat.com>
359 for reporting this bug.
360
361 5. Made restore build on Solaris, making possible to
362 restore Linux's "enhanced" tapes. Thanks to Uwe Gohlke
363 <uwe@ugsoft.de> for the patch.
364
365 6. Made an extension in the dump tape format capable of saving
366 MacOSX specific inode extensions. Uwe Gohlke <uwe@ugsoft.de>
367 wrote the extension and contributed the restore code back
368 into this codebase. The same extension mechanism will be
369 used in the future to save ACLs...
370
371 7. Made rmt work correctly with regard to QFA and local/remote
372 files and tapes. The remote access will however work only
373 when the dump provided rmt version is used. If you want to
374 use another rmt server, please do not use the QFA feature.
375 Thanks to Uwe Gohlke <uwe@ugsoft.de> for the patch.
376
377 Changes between versions 0.4b33 and 0.4b34 (released April 18, 2003)
378 ====================================================================
379
380 1. Fixed the INSTALL file to reflect the actual install paths.
381 Thanks to David Ronis <ronis@ronispc.chem.mcgill.ca> for
382 reporting the bug.
383
384 2. Fixed the configure script to only check for headers presence
385 instead of trying to compile them. This should fix issues
386 with old build environments. Thanks to Kari Lempiainen
387 <kari@funky.pp.fi> for reporting the bug.
388
389 3. Fixed restore to correctly ignore sockets when comparing
390 a dump (as socket cannot be properly restored anyway). Thanks
391 to Gunther Reiszig <gunther@mit.edu> for reporting the bug.
392
393 4. Fixed restore to correctly access the archive file (-A argument)
394 even when using a remote tape. Thanks to Fabrice Bellet
395 <fabrice@bellet.info> for reporting the bug.
396
397 5. Fixed (again) handling of long (largefile) seeks in rmt.
398 Thanks to Fabrice Bellet <fabrice@bellet.info> for reporting
399 the bug.
400
401 6. Fixed restore corner case when dealing with large block sizes
402 dump is able to write now (-b 1024). Thanks to Fabrice Bellet
403 <fabrice@bellet.info> for reporting the bug.
404
405 7. Fixed a bug preventing dump to access a filesystem having
406 a label exactly 16 bytes in length. Thanks to <minduni@ti-edu.ch>
407 for reporting the bug.
408
409 8. Made dump store dump dates using explicit timezones, fixing a
410 problem with incremental dumps if the timezone is modified
411 between the dumps. Thanks to Philippe Troin <phil@fifi.org> for
412 the bug report and the patch.
413
414 9. Fixed a bug encountered when dumping individual files (not full
415 filesystems or directories) and dangling symbolic links happen
416 to be in the list of files. For as far as dump is concerned,
417 dangling symbolic links are allowed, and are dumped as is.
418 Thanks to Jin-su Ahn <jsahn@ee.snu.ac.kr> for reporting the
419 bug and providing the fix.
420
421 10. Fixed open and creation modes and permissions for QFA and
422 table-of-contents files in dump and restore. Thanks to
423 Philippe Troin <phil@fifi.org> for the patch.
424
425 11. Fixed the archive file descriptor handling enabling it to be 0.
426 This can happen in some cases when shell redirections are used.
427 Thanks to Philippe Troin <phil@fifi.org> for the patch.
428
429 12. Delayed the opening of archive file until after suid had been
430 dropped (fixing a possible security issue if dump is suid).
431 Thanks to Philippe Troin <phil@fifi.org> for the patch.
432
433 13. Fixed the 'S' command handling in the rmt client part.
434 Thanks to Philippe Troin <phil@fifi.org> for the patch.
435
436 14. Modified the end-of-tape script handling to print out statistics
437 (and stop the timer) before launching the eot script. Also, the eot
438 script does not get run anymore when using -M (which makes sense) or
439 when multiple tapes are listed on the command line
440 (-f tape0,tape1,tapen) (which also makes sense).
441 Thanks to Philippe Troin <phil@fifi.org> for the patch.
442
443 15. Relicensed dump/restore under the 'revised' BSD license, as per
444 ftp://ftp.cs.berkeley.edu/ucb/4bsd/README.Impt.License.Change.
445
446 16. Added LZO compression to dump. This new compression method has
447 the advantage of being super fast, thus not killing tape streaming
448 on slow machines. Thanks to Helmut Jarausch
449 <jarausch@igpm.rwth-aachen.de> for the patch and to
450 Markus Oberhumer <markus@oberhumer.com> for giving special permission
451 to include his miniLZO project (GPL licensed) in dump/restore.
452
453 17. Some small buffer overruns fixes in rmt. Thanks to Antonomasia
454 <ant@notatla.demon.co.uk> for reporting the bugs.
455
456 18. Added a special rmt version which can do encryption when writing
457 to tape. Read examples/encrypted_rmt/README for details on
458 how to enable and configure it. Thanks to Ken Lalonde
459 <ken@globalremit.com> for the patch.
460
461 19. Made dump work with 2.5 kernel end of tape early warning semantics.
462 Thanks to Kai Makisara <Kai.Makisara@kolumbus.fi> for the patch.
463
464 20. Fixed a bug which caused dump -w|-W not to work anymore, because
465 the fs_freq and fs_passno fields in /etc/mtab are always set
466 to 0 0. Thanks to Trent Piepho <xyzzy@speakeasy.org> for
467 reporting the bug.
468
469 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
470 =======================================================================
471
472 1. Added a note in the restore man page clarifying the question
473 'set the permissions on the current directory ?' asked by
474 restore at the end of treatment in -i and -x modes.
475
476 2. Fixed the endianess issues when reading compressed tapes.
477 Thanks to Dark Force <daq4th@users.sourceforge.net> for
478 reporting this bug and providing test cases.
479
480 3. Fixed the "ACL won't be dumped" warning message (which showed
481 an extra, unrelated error message). Thanks to Dragan Krnic
482 <dkrnic@lycos.com> for reporting this bug.
483
484 4. Made dump look first into /etc/mtab, then into /etc/fstab
485 when searching for filesystem paths. Also fixed some problems
486 caused by binding mounts or multiple block device mounts.
487 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
488 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
489 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
490 and #160305, Sourceforge bugs #618699 and #679832).
491
492 5. Made dump's -I option accept the value '0' meaning all the
493 read errors will be ignored. This can be useful when running
494 dump from unattended sessions (like cron jobs). Thanks to
495 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
496
497 6. Fixed the output of dump to indicate 'blocks' instead of
498 'tape blocks' in the various messages (blocks are always
499 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
500 made it clearly print the current blocksize at the start of
501 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
502 the suggestions.
503
504 7. Made rmt understand long (largefiles) seeks.
505
506 8. Fixed build with very old versions of libext2fs, where
507 EXT2_FT_* constants were undefined.
508
509 9. Made dump accept the dumpdates path on the command line
510 (-D file option) instead of using only the hardcoded one.
511 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
512 suggestion.
513
514 10. Enabled rmt, LFS, readline, QFA options by default in
515 ./configure. Updated the configure process (new versions
516 of config.guess, config.sub etc).
517
518 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
519 =======================================================================
520
521 1. Changed dump to use fcntl(F_SETLK) style locking instead
522 of flock() when locking the dumpdates file. With the old
523 locking scheme, a local user having read rights on the
524 dumpdates file could be able to do a Denial of Service attack
525 on dump. In order to lock the dumpdates file with the new
526 scheme, the user would need to have write access on the file.
527 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
528 reporting the bug (originally a bugtraq post).
529
530 2. Fixed interactive 'ls' which caused spurious errors warnings
531 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
532 <jorgen_ostling@users.sourceforge.net> for reporting this
533 bug.
534
535 3. Fixed dump's estimate when dealing with sparse inodes.
536
537 4. Modified dump to allow setting a blocksize bigger than 32kB
538 (raised the limit to 1024kB), because newer hardware needs
539 this for proper operation. Thanks to Dirk Traenapp
540 <dtraenapp@users.sourceforge.net> for reporting this.
541
542 5. Fixed a bug causing Dump to stop and report an error if an
543 inode number in the exclude file was followed by some amount
544 of whitespace. Thanks to Jeffrey Sofferin
545 <sofferin@users.sourceforge.net> for reporting this bug.
546
547 6. Fixed a bug which caused restore, in some particular cases,
548 to ask some 'scary' questions and leave a bunch of RSTTMP
549 directories behind when restoring incremental tapes. Thanks
550 to Philippe Troin <phil@fifi.org> for reporting this bug and
551 providing the test cases.
552
553 7. Changed the wording when inodes are excluded from dump:
554 replaced 'Added inode 7 to exclude list' with
555 'Excluding inode 7 (resize inode) from dump', as suggested
556 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
557
558 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
559 ===================================================================
560
561 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
562 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
563 <eros@lamel.bo.cnr.it> for reporting the bug.
564
565 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
566 ===================================================================
567
568 1. Made dump print out the ext2 logical block number in case of a read
569 error, which can be given as an argument to debugfs' ncheck command,
570 in order to find out the affected inode. Added note about this
571 usage in dump's man page.
572
573 2. Fixed a problem in restore when reading tapes written on big
574 endian machines with very old versions of dump. The patch was
575 contributed by George Helffrich <george@geology.bristol.ac.uk>.
576
577 3. Fixed the tape length calculation when using large tapes
578 and compression. Thanks to Georg Lippold
579 <g_lippold@sourceforge.net> for reporting the bug.
580
581 4. Added a new set of examples in dump_on_cd_2 directory, based
582 on dump_on_cd examples but somewhat enhanced, supporting DVD
583 media, and localized in english and german. Thanks to
584 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
585
586 5. Made dump save 32 bit UID/GID. Internally, this was achieved
587 by switching from the old BSD inode format to the new BSD
588 inode format, which means that the tape format was changed.
589 However, since all restore versions out there should
590 transparently support both inode formats, the change should
591 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
592 reporting the bug.
593
594 6. Fixed a lot of warnings in the code shown when compiling
595 with 'gcc -W'. Thanks to Matthias Andree
596 <matthias.andree@stud.uni-dortmund.de> for reporting this.
597
598 7. Fixed a small markup bug in the dump man page. Thanks to
599 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
600 patch.
601
602 8. Rewrote entirely the man pages using the tmac.an macro
603 package (Linux man page format) instead of the original BSD
604 format. They should be now cleaner and easier to modify.
605
606 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
607 ==================================================================
608
609 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
610 non Linux clients were misinterpreted. The description of the problem
611 (incompatible numbering in Linux mtio opcodes) is documented at
612 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
613 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
614 providing an excellent, cross-platform replacement for rmt in his
615 star package.
616
617 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
618 which caused restore to fail when the inode maps were not entirely
619 contained on the first volume (when using really small volumes or
620 when dumping a huge number of inodes).
621
622 3. Fixed a problem in dump, where files in subdirectories of directories
623 that have the nodump flag set (or being excluded with -e/-E) were
624 sometimes incorrectly dumped (instead of being excluded). The original
625 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
626 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
627 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
628 help testing the patch.
629
630 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
631 which makes one able to pipe the output of dump, by the net, to
632 a remote CD-burner server.
633
634 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
635 and use GNU's symbolic syntax over rmt instead of numerical values
636 to assure multiple platform compatibility.
637
638 6. Documented the -d option in restore.
639
640 7. Added a -v (verbose) mode to dump. For now it just prints the number
641 of the inode being dumped, but this could evolve in future versions
642 to include interesting debugging output.
643
644 8. Added a -o flag to restore, which automatically restores the current
645 directory permissions in -i or -x mode, without asking the operator
646 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
647 and Peter Samuel <peters@e-smith.com>.
648
649 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
650 ====================================================================
651
652 1. Fixed a bug in the QFA feature which could have caused
653 a bad tape positionning on restore (causing extra delays in
654 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
655
656 2. Added a small note in the dump man page specifying that
657 there should be no white space between the option letter and
658 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
659
660 3. Made restore work with older versions of the readline library,
661 by checking in configure for several extended symbols. Restore
662 can now be compiled with a readline as old as the 2.0 release
663 (though it may be a good idea to upgrade it to a more recent
664 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
665 reporting the build failures.
666
667 4. Fixed a performance problem with the QFA file creation in
668 dump, which made unnecessary seeks on the tape slowing down
669 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
670 reporting this issue.
671
672 5. Removed the inclusion of some kernel headers in the dump
673 source, which prevented the compile in some kernel/glibc
674 headers/architecture combination. Thanks to Bdale Garbee
675 <bdale@gag.com> for reporting the bug.
676
677 6. Added the appropriate error message when dump fails to
678 open the output file for writing. Thanks to Amith Varghese
679 <amithv@yahoo.com> for reporting this bug.
680
681 7. Made restore able to understand large Solaris ufsdump tapes
682 (containing inodes bigger than 4194304). Sun have introduced
683 an "extension" to the dump tape format when dealing with
684 those inodes, which was uncorrectly handled by Linux restore.
685 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
686 providing a test case.
687
688 8. Added the -m parameter to dump which optimises the output for
689 inodes having been changed but not modified since the last dump
690 ('changed' and 'modified' have the meaning defined in stat(2)).
691 For those inodes, dump will save only the metadata, instead of
692 saving the entire inode contents. Inodes which are either
693 directories or have been modified since the last dump are saved
694 in a regular way. Uses of this flag must be consistent, meaning
695 that either every dump in an incremental dump set have the flag,
696 or no one has it.
697
698 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
699 =======================================================================
700
701 1. Fixed behaviour of dump when exceeding resource limits
702 (SIGXFSZ treatment).
703
704 2. Added the -L flag to restore to allow the user to specify a
705 maximal allowed number of miscompares when using restore
706 with the -C option to check the backup.
707
708 3. Detailed the manual entry for the -N option of restore.
709
710 4. Added the -a flag to restore to make able doing unattended
711 restores in -i or -x mode (automatically walks through the
712 multiple dump volumes).
713
714 5. Extended the QFA mode to work with local files and/or
715 remote tapes and files. This way, restore can know in advance
716 the tape number and the offset for the inodes to extract and
717 can minimize the extraction time by seeking directly to the
718 good tape/offset.
719
720 6. Added the -A <archive> option to both dump and restore,
721 which makes dump to archive a dump table-of-contents in
722 the specified file to be used by restore to determine
723 whether a file is in the dump file that is being restored.
724 (the archive file syntax is also compatible with the
725 Solaris ufsdump generated one).
726
727 7. Small fix in restore making it able to read some (broken ?)
728 Solaris ufsdump tapes.
729
730 8. Fixed dump to correctly recognise the root filesystem when using
731 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
732 for reporting this bug.
733
734 9. Added the -P <file> option to restore to create a
735 Quick File Access file from an already made dump. Patch
736 contributed by Uwe Gohlke <uwe@ugsoft.de>.
737
738 10. Made restore compile and run on Solaris, making it a
739 possible replacement for the standard ufsrestore. Port was
740 contributed by Uwe Gohlke <uwe@ugsoft.de>.
741
742 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
743 =====================================================================
744
745 1. Added a set of backup scripts from Eugenio Diaz
746 <getnito@yahoo.com> in the examples section. It features
747 automatic (cron based) full and incremental dumping of
748 several filesystems on a separate filesystem.
749
750 2. Fixed a off-by-one miscalculation which disabled dumping
751 a one letter subdirectory. Thanks to Chris Siebenmann
752 <cks@utcc.utoronto.ca> for reporting the bug.
753
754 3. Fixed several restore bugs occuring when trying to
755 restore some missing files on the tape. Thanks to Chris
756 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
757
758 4. Fixed --with-ldopts configure argument passing, installing from
759 a separate object directory, makefile cleanups contributed
760 by <splite@purdue.edu>.
761
762 5. Fix a bug which could caused, in some conditions, the highest
763 number inode of a filesystem, to not be dumped. Many thanks
764 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
765 chase this bug.
766
767 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
768 =======================================================================
769
770 1. Added a mini howto from Patrick Walsh in the examples
771 directory.
772
773 2. Minor man pages syntax corrections. Thanks to
774 Chris Danis <screechco@home.com> for reporting the bugs.
775
776 3. Added a script from David B. Peterson <dave@toppledwagon.com>
777 to the examples section. It features dumping several
778 filesystems to a remote tape drive upon ssh.
779
780 4. Added a patch provided by Richard Jones <rich@annexia.org>
781 which allows BRADEMAX (number of read errors tolerated by
782 dump) to be adjusted using the -I option.
783
784 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
785 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
786 patch.
787
788 6. Add the -l option to restore to specify if, when doing a
789 remote restore, the file used to access the backup is a
790 regular file (the defaults being a tape drive). Restore needs
791 to know this information when reading a remote compressed
792 dump. Previously, this information was autodetected, but
793 the autodetection code fails (with ioctl: Inappropriate ioctl
794 for device) when using a non Linux remote box. Thanks to
795 many users and especially to Eros Albertazzi
796 <eros@lamel.bo.cnr.it> for reporting this.
797
798 7. Found a workaroung for the dump deadlock problem (3 childs
799 stuck in pause(), father in read()). The workaround seems
800 to work for me and several beta-testers. If it doesn't work
801 for you, please report back.
802
803 8. Updated the RPM spec file (BuildPrereq, URL etc).
804
805 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
806 ========================================================================
807
808 1. Fixed the permissions of a newly created QFA file by dump.
809
810 2. Cleaned up the source of dump (the external variables
811 definition was a complete mess, making possible to have
812 objects overlap).
813
814 3. Fixed restore to use the full tape volume path when doing
815 a compare (since it changes the working directory to the
816 filesystem being compared in the process).
817
818 4. Added the -q option to dump which makes dump abort
819 whenever it needs operator attention. It should be
820 easier to use dump in scripts with this option.
821
822 5. Detect the use of incompatible options to dump and
823 refuse them (like -a and -B options together).
824
825 6. Added bzip2 compression to dump/restore (use option -j level
826 to select it). Note that you will need a restore version
827 >= 0.4b24 in order to restore a bzip2 compressed dump.
828 The same warning as for the zlib compression applies:
829 the tape format of a bzip2 dump is not compatible with the
830 original BSD tape format.
831
832 7. Fixed a overflow problem in dump corrupting the dump when
833 very large files were encountered. Thanks to Vladimir Nadvornik
834 for the bug report.
835
836 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
837 kernel buffer/page cache before starting the dump, helping
838 a bit those who use dump on mounted filesystems. Thanks to
839 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
840 for this suggestion.
841
842 9. Updated the RPM spec file following the RedHat changes
843 (dynamically linked binaries now in /usr/sbin etc).
844
845 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
846 which enables restore to recognise multi volume compressed dumps
847 done on CD. Included his scripts for dump (which pipe the dump
848 output directly into cdrecord) and restore. There is now
849 possible to to backups to CD on the fly!
850
851 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
852 ===================================================================
853
854 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
855 Marcus Meissner (Caldera International Security Dept.).
856
857 2. Implement the Sun rmt extended protocol. Patch submitted
858 by Ian Gordon <iangordon@users.sourceforge.net>.
859
860 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
861 the linux kernel header. This ensures that dump always has the
862 latest version of this file. Patch submitted by
863 Andreas Dilger <adilger@turbolinux.com>.
864
865 4. Report any filesystem present in either /etc/fstab with a
866 positive passno or /etc/dumpdates in dump -w output.
867 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
868
869 5. Fixed the looping problem in dump introduced in the
870 previous version.
871
872 6. Changed the -B option of dump to limit the size of
873 _compressed_ output per volume if compression is on.
874 Patch contributed by Helmut Jarausch
875 <jarausch@igpm.rwth-aachen.de>. Note however that, since
876 it is impossible to predict the size of the compressed
877 data before doing the compression, the -B limit is a bit
878 conservative.
879
880 7. Fixed a bug in reading the operator typed file/tape path for
881 the next volume in restore.
882
883 8. Implemented a "-F script" option to restore which permits the
884 user to specify a script which will be launched at the
885 beginning of each tape, useful for automatic programming of
886 tape changers for example. See the restore man page for the
887 script parameters and return codes.
888
889 9. Small fix for the QFA routines provided by Uwe Gohlke
890 <uwe@ugsoft.de>, and some recommendations for QFA uses in
891 the man pages.
892
893 10. Fixed the multivolume restoring where making a mistake
894 to the 'Mount next tape' prompt caused several blocks to
895 be lost.
896
897 11. Enhanced the -e option of dump to take as a parameter a
898 comma separated list of inode numbers.
899
900 12. Added the -E option to dump which specify a file containing
901 inode numbers to exclude from the dump.
902
903 13. Fixed the compressed multi-volume dump + restore.
904
905 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
906 ==================================================================
907
908 1. Made dump report the number of blocks dumped per volume.
909 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
910
911 2. Fix a bug in dump where waiting too much at the 'change volumes'
912 question caused the volume to be rewritten. Thanks to
913 Graham Reed <greed@users.sourceforge.net> for reporting the
914 bug and providing a patch.
915
916 3. Added a compression option to dump, contributed by Jerry
917 Peters <gapeters@worldnet.att.net>.
918
919 WARNING: the compression works only on files/pipes and
920 drives supporting variable block size.
921
922 WARNING2: tapes created using the compression feature are
923 incompatible with the standard dump tape format, and a
924 version of restore >= 0.4b22 will be needed for extraction.
925
926 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
927 architectures. Thanks to Paul Slootman <paul@debian.org> for
928 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
929 it upstream.
930
931 5. Many cleanups (CPP defines, const char warnings, check of
932 ext2fs COMPAT flags, time_t cleanups, added libext2 version
933 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
934
935 6. Made --prefix option work in configure. All the install path
936 are now based on the configure parameters.
937
938 7. Added the Quick File Access mode in dump/restore, contributed
939 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
940 a file tape position for each inode, and this file is used by
941 restore (if called with parameter Q and the filename)
942 to directly position the tape at the file restore is currently
943 working on. This saves hours when restoring single files from
944 large backups, saves the tapes and the drive's head. Use
945 --enable-qfa option of configure to compile in the QFA support.
946
947 8. Added the possibility to dump several files and directories
948 in a single invocation of dump. Thanks to Uwe Gohlke
949 <uwe@ugsoft.de> for implementing this option.
950
951 9. Fixed the dumping and restoring of files with holes
952 (files ending with holes didn't get dumped or restored
953 correctly).
954
955 10. Fixed a socket descriptor leak in dump, which leaved opened
956 3 file descriptors per dump process (and there is one dump
957 process per tape).
958
959 11. Fixed dump large file system support, by explicit use of
960 open64/lseek64/etc functions (explicit use needed because
961 e2fsprogs libraries don't behave well when compiled with
962 FILE_OFFSET_BITS=64).
963
964 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
965 ======================================================================
966
967 1. Fixed some bugs in the dump -w|-W logic introduced by
968 the previous version. Thanks to Andreas Dilger
969 <adilger@turbolinux.com> for his help on this one.
970
971 2. Fixed again a compilation problem when using old e2fs
972 headers (filesystem label related). Thanks to many users
973 who reported this stupid error.
974
975 3. Fixed a build problem on old lib5 based systems dealing with
976 _PATH_MNTTAB being undefined. Thanks to John Adams
977 <johna@onevista.com> for reporting the bug.
978
979 4. Improved the error detection in restore when creating
980 the temporary files in TMPDIR. Now restore will corectly
981 report a 'no space left on device' error instead of
982 strange errors which could imply an archive corruption.
983 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
984 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
985
986 5. Added the throughput information in dump's progression
987 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
988 for the patch.
989
990 6. Use libext2fs's inode scanning routines, which are particularly
991 robust when dealing with errors (especially when having some
992 bad blocks on the filesystem). Many thanks to Colin
993 <colin@horizon.com> for reporting the bug, investigating
994 the issues, finding a workaround, writing the patch and
995 fully testing it... (of course, if this patch will break
996 anything, he is to blame for :-)).
997
998 7. Made dump and restore LFS aware. Dump can dump a filesystem
999 containing large files, generate a large file on output and
1000 restore can restore them. This feature is not enabled by
1001 default, use --enable-largefile option of configure to enable
1002 it (you need to have a LFS aware glibc though). Thanks to
1003 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
1004 and to Theodore T'so <tytso@valinux.com> for his always
1005 useful thoughts.
1006
1007 8. Made dump ask upon a tape write error if it should rewrite
1008 the current volume (assume this is a bad tape) or if it should
1009 assume an end-of-tape condition (useful for tape drives which
1010 behaves badly at the end of the tape). Thanks to Andreas
1011 Dilger <adilger@turbolinux.com> for the suggestion.
1012
1013 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
1014 =======================================================================
1015
1016 1. Fixed a small compilation problem due to a change
1017 in the definintion of the struct sigaction in
1018 glibc 2.0 / libc5. Thanks to Gunther Schlegel
1019 <schlegel@riege.de> for reporting the bug and to
1020 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
1021 a fix.
1022
1023 2. Modified the label and uuid analysis in order to be
1024 self-contained (independant of kernel/libc headers). This
1025 should fix the compile with older kernel/libc headers and
1026 will preserve the functionality. Thanks to Bernhard Erdmann
1027 <bernhard.erdmann@gmx.de> for reporting the bug.
1028
1029 3. The 'exclude inode' option, if applied to a directory
1030 inode, excludes now correctly all the leaves of this
1031 directory. Thanks to John R. Dennison
1032 <gerdesas@users.sourceforge.net> for reporting the bug.
1033
1034 4. Fixed the '-e' option to disable the possibility
1035 to exclude the root inode (which causes the dumps to
1036 be unreadable by restore). Prevented array overflow
1037 when multiple -e options are used.
1038
1039 5. Fixed dump to correctly interpret a filesystem argument
1040 which is a mountpoint and it is not an absolute path
1041 (as specified in the fstab). Thanks to Bernhard R. Erdmann
1042 <be@berdmann.de> for reporting the bug.
1043
1044 6. Made dump able to backup files larger than 2 GB. Note that
1045 dump still doesn't cope with files larger than 4 GB.
1046
1047 7. Restore the real uid and gid before invoking an external
1048 RSH program (big hole when dump or restore is suid root!).
1049
1050 8. Get the values for the default tape device and the location
1051 of fstab file from the system headers. Thanks to
1052 Andreas Dilger <adilger@turbolinux.com> for the patch.
1053
1054 9. Made dump -w|-W report all recognized filesystems
1055 present in either /etc/fstab or /etc/dumpdates, and present
1056 the list in the same order as in fstab file. Thanks
1057 to Andreas Dilger <adilger@turbolinux.com> for the patch.
1058
1059 10. Made dump's -a (automatic end of tape detection) the
1060 default. Specifying one of -B, -c, -d or -s options will
1061 override the EOM detection. Thanks to Andreas Dilger
1062 <adilger@turbolinux.com> for the patch.
1063
1064 11. Save the ext2 filesystem label into the dump volume label.
1065 Specifying a volume label on the command line (-L option)
1066 will override this feature. Thanks to Andreas Dilger
1067 <adilger@turbolinux.com> for the patch.
1068
1069 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
1070 =====================================================================
1071
1072 1. Fixed the signal handling in dump (which I broke in 0.4b17)
1073 which was causing several strange dump failures (dump
1074 hanged or segmentation faults etc).
1075
1076 2. Specified the default blocksize in the dump man page.
1077
1078 3. Changed two info messages of restore to be written on stdout
1079 instead of stderr in order to leave stderr only for errors
1080 or warnings. Thanks to Stephen Atwell
1081 <satwell@urbana.css.mot.com> for the suggestion.
1082
1083 4. Corrected an off by one calculation which prevented
1084 dumping a directory having a 1 character name length.
1085 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
1086 for reporting the bug.
1087
1088 5. Reinforce dump in order to allow only level 0 and no
1089 -u option when dumping a subdirectory, and document
1090 this in the man page. Thanks to Bernhard Erdmann
1091 <bernhard.erdmann@gmx.de> for reporting the bug.
1092
1093 6. Fixed a small harmless bug in restore which caused
1094 in some conditions hard links to be restored several
1095 times (therefore generation some warning messages).
1096 Thanks to Kenneth Porter <shiva@well.com> for
1097 reporting the bug.
1098
1099 7. Updated the RPM spec file to the latest RedHat version,
1100 providing FHS packaging and other cosmetic changes.
1101 You will need RPM version >= 3.0.5 in order to build the RPM.
1102
1103 8. Updated the configure script to check for libtermcap
1104 before checking for libreadline, because we need this
1105 library in order to compile the readline support.
1106
1107 9. Made dump understand the LABEL= and UUID= notation
1108 both in /etc/fstab and on the command line. Note that
1109 you will need the /proc filesystem in order to use
1110 these notations. Thanks to Erik Troan <ewt@redhat.com>
1111 for providing the patch.
1112
1113 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
1114 ===================================================================
1115
1116 1. Fixed a potential buffer overflow in restore. Thanks
1117 to Stan Bubrouski <satan@fastdial.net> for reporting
1118 the bug.
1119
1120 2. Fixed a readline-related bug which prevented
1121 'cat DUMPFILE | restore -i -f -' from working. Thanks
1122 to Charles Karney <karney@users.sourceforge.net>
1123 for the bug report.
1124
1125 3. Changed a few "panic" into "exit", causing restore to
1126 be more stable against some attacks (like the last one
1127 reported on Bugtraq, although the last version of restore
1128 was not vulnerable - just dumped core). Thanks to
1129 Andreas Hasenack <andreas@conectiva.com.br> for reporting
1130 the bugs to me.
1131
1132 4. Removed the suid-root bit on dump and restore in the
1133 default build (and generated RPMs). It should be safer
1134 now. Users who need the suid-root capabilities in order
1135 to make network backups should read first the man page
1136 of dump and enable the suid bit manually.
1137
1138 5. Added -ltermcap to the compile parameters for restore
1139 when using readline support, in order to make the compile
1140 process work on some systems (SuSE ?). Thanks to
1141 Patrik Schindler <poc@pocnet.net> for reporting the bug.
1142
1143 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
1144 ====================================================================
1145
1146 1. The -F script is called now *only* at the end of a tape,
1147 not at the end of the dump. Thanks to Christian Haul
1148 <haul@informatik.tu-darmstadt.de> for the bug report.
1149
1150 Normally, the device name and the current volume number
1151 are passed on the command line of the script. However,
1152 if you want the old style script (with no arguments
1153 passed on the command line), you can enable it in
1154 configure with the --enable-oldstylefscript.
1155
1156 2. Use posix signal handling to preserve dump functionality
1157 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
1158 the patch.
1159
1160 3. Made the exit code of restore in case of a 'C'ompare
1161 command reflect the result of the compare. An exit status
1162 of 0 means the dump archive is correct, 1 means tape errors,
1163 2 means that some files were modified. Thanks to Kenneth Porter
1164 <shiva@well.com> for the suggestion.
1165
1166 4. Made (finally) quotes work in interactive restore.
1167
1168 5. Small fixes in order to allow dump to compile with a
1169 really old e2fsprogs version. Thanks to Ian Zimmerman
1170 <itz@speakeasy.org> for the bug report.
1171
1172 6. Add GNU readline capabilities to interactive restore.
1173 Use configure's --enable-readline flag to enable this feature.
1174 Thanks to Patrik Schindler <poc@pocnet.net> for the
1175 suggestion.
1176
1177 7. Do the compare on the fly in restore 'C' mode (this will
1178 allow not to exhaust the available /tmp space when
1179 ccmparing large files). Thanks to Kenneth Porter
1180 <shiva@well.com> for the suggestion.
1181
1182 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
1183 ====================================================================
1184
1185 1. Fixed some several duplicate 'DUMP: DUMP:' in the
1186 output of dump.
1187
1188 2. Corrected the estimation of blocks to dump. Note that
1189 this is still wrong for large directory inodes, due
1190 to the size of a BSD directory entry on the tape
1191 (estimation is based on the size of the ext2 directory,
1192 which is smaller because it doesn't contain the null
1193 character at the end).
1194
1195 3. Made dump report the total number of MB written to
1196 tape at the end of run. Thanks to W. Reilly Cooley
1197 <wcooley@nakedape.cc> for the patch.
1198
1199 4. Added the -X option to restore giving the possibility
1200 to read the names of the files to be extracted/listed
1201 from a text file (in addition of the command line).
1202 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
1203 patch.
1204
1205 5. Added the device name and the current volume number
1206 as arguments to the end of tape script (dump -F option).
1207
1208 6. Made the multi-volume dump work again (after having
1209 broken it in 0.4b15).
1210
1211 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
1212 ===================================================================
1213
1214 1. Added a prompt command in interactive restore mode. Thanks
1215 to Andreas Dilger <adilger@home.com> for the patch.
1216
1217 2. Fixed a buffer overflow problem in dump (caused by
1218 not checking the size of the filesystem parameter).
1219 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
1220 reporting this on Bugtraq (and to several dump users
1221 who forwarded me his mail).
1222
1223 3. Added the '-F script' option to dump in order to
1224 launch a script at the end of each tape (to be used
1225 with a tape changer, or to notify the sysadmin by
1226 pager etc.).
1227
1228 4. Fixed a bug in restore compare code caused by the changes
1229 I made in 0.4b14.
1230
1231 5. Fixed the treatment of options using the old BSD syntax
1232 in both dump and restore.
1233
1234 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
1235 =======================================================================
1236
1237 1. Fixed a bug in dump which may have caused invalid deleted
1238 directories to be dumped out if they were deleted after the
1239 mapping pass. This could occure on active filesystem and lead
1240 to heap corruption (causing dump malfunction in many possible ways).
1241 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
1242 bug and submitting the patch.
1243
1244 2. Fixed the handling of the filesystem parameter in dump. A
1245 '/mnt/disk' parameter caused the disk contents to be dumped,
1246 but a '/mnt/disk/' parameter caused the mountpoint directory
1247 to be dumped (generally an empty directory).
1248
1249 3. Improved the output of dump in order to tell which directory
1250 it is currently dumping (when dumping a subtree).
1251
1252 4. Added the '-e' exclude inode option to dump. Thanks to
1253 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
1254
1255 5. Added a REPORTING-BUGS file in order to provide a guide
1256 on how to correctly report a bug in dump/restore.
1257
1258 6. Fixed a restore problem when restoring a hard link to an inode
1259 having the immutable or append-only attribute set. Thanks to
1260 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
1261
1262 7. Fixed a compatibility problem between dumps made on little
1263 endian machines (the format was correct) and big endian
1264 machines (the format was incorrect). This fix break the
1265 compatibility with the older dumps made on big endian
1266 machines (sparc, m86k, ppc etc). For the first time in
1267 linux dump's history (I believe), the dumps made by *BSD,
1268 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
1269 so interchangeable. Thanks to Rob Cermak
1270 <cermak@ahab.rutgers.edu> for submitting the bug and
1271 helping me test the fix.
1272
1273 8. Fixed the way dump reports the remaining percent/time, if
1274 the number of blocks actually dumped exceeds the estimated
1275 number of blocks. Thanks to Jean-Paul van der Jagt
1276 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
1277
1278 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
1279 ======================================================================
1280
1281 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
1282 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
1283
1284 2. Removed the requirement to build the RPM as root from the
1285 spec file. Thanks to Christian Weisgerber
1286 <naddy@mips.rhein-neckar.de> for submitting this.
1287
1288 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
1289 causing dump to block when dumping really huge filesystems.
1290 Many thanks to Patrik Schindler <poc@pocnet.net> for
1291 helping me find this bug.
1292
1293 4. Fixed the treatment for an interrupt signal when dump access
1294 the remote tape through RSH. Thanks to Christian Weisgerber
1295 <naddy@mips.rhein-neckar.de> for providing the patch.
1296
1297 5. Fixed a bug which was causing dump/restore to display
1298 garbage characters instead of the remote host name.
1299
1300 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
1301 =====================================================================
1302
1303 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
1304 <kukuk@suse.de> for submitting the patch.
1305
1306 2. Fix for the exit code when using the size estimate option of
1307 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
1308 submitting the patch.
1309
1310 3. Handle EINTR in atomical reads/writes in dump, which was causing
1311 dump to fail on some systems. Thanks to Eric Jergensen
1312 <eric@dvns.com> for reporting the bug and submitting the patch.
1313
1314 4. Handle more than 16 characters for the device names in dumpdates.
1315 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
1316 tracking down the problem and proposing the solution.
1317
1318 5. Fixed a bug in dump which prevented the creation of the
1319 dumpdates file when doing a 0-level dump without already
1320 having a dumpdates file. Thanks to Patrik Schindler
1321 <poc@pocnet.net> for reporting the bug.
1322
1323 6. Changed the way dump 'S' flag reports the size estimate
1324 from number of blocks into bytes (making it compatible
1325 with the Solaris version, and simplifying things for
1326 amanda users). Thanks to Jason L Tibbitts III
1327 <tibbs@math.uh.edu> for reporting the bug.
1328
1329 7. Fixed a compatibility problem in linux/alpha dump tape format.
1330 Now the linux/alpha dump are (again) compatible with the
1331 other dump formats. But this breaks compatibility with
1332 older dumps made on alpha. Thanks to Mike Tibor
1333 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
1334
1335 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
1336 ======================================================================
1337
1338 1. Added a '--enable-kerberos' to configure.
1339
1340 2. Added a 'S' option to dump which determines the amount of space
1341 that is needed to perform the dump without actually doing it, similar
1342 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
1343 <cermak@ahab.rutgers.edu>.
1344
1345 3. Added a 'M' multi-volume option to dump and restore which enables
1346 dumping to multiple files (useful when dumping to an ext2
1347 partition to make several dump files in order to bypass the 2GB
1348 file size limitation). The 'f' argument is treated as a prefix and
1349 the output files will be named <prefix>001, <prefix>002 etc. With
1350 the 'M' flag, restore automatically selects the right file without
1351 asking to enter a new tape each time.
1352
1353 4. Fixed a memory leak which was causing dump to grow very big
1354 (270MB when dumping a 10GB filesystem...). Thanks to Jason
1355 Fearon <jasonf@netrider.org.au> for reporting the bug.
1356
1357 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
1358 ======================================================================
1359
1360 1. Make configure test if the system glob routines support
1361 extended syntax (ALTDIRFUNC). If not, use the internal glob
1362 routines instead of system ones. Thanks to Bernhard Sadlowski
1363 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
1364 and helping me resolve this and other minor libc5 compiling
1365 glitches.
1366
1367 2. Fix a problem when dumping a ext2fs with the 'filetype'
1368 feature enabled. Thanks to Patrick J. LoPresti
1369 <patl@cag.lcs.mit.edu> for reporting the bug and to
1370 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
1371
1372 3. Made the nodump flag work on directories. A directory which
1373 has the nodump flag gets never dumped, regardless of its
1374 contents.
1375
1376 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
1377 which allows dump on an active ext3 filesystem. However, this
1378 is a "quick and dirty" patch which enables backup of an ext3
1379 filesystem through the ext2 compatibility (by ignoring the
1380 NEEDS_RECOVERY bit). The journal file is not recognized and
1381 it is dumped (it should not).
1382
1383 5. Test the superblock compatibility flags when dumping, in order
1384 to be sure that we know how to deal with specific features.
1385
1386 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
1387 ====================================================================
1388
1389 1. Use lchown instead of chown, fixing a possible security problem
1390 when restoring symlinks (a malicious user could use this
1391 to deliberately corrupt the ownership of important system files).
1392 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
1393 this and providing the patch.
1394
1395 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
1396 ====================================================================
1397
1398 1. Put dump sources under CVS, added Id tags in all files so
1399 one can use 'ident' on binary files.
1400
1401 2. Added the dump/restore version in the usage text so one can
1402 easily verify the version he is using.
1403
1404 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
1405 a va_start/va_end problem on linux-ppc (always call va_start
1406 va_end in pairs each time we use a vararg function).
1407
1408 4. Added again the DT_* constants because old libc does not
1409 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
1410 for submitting the bug report.
1411
1412 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
1413 this should enable dumping big (huge) filesystems.
1414
1415 6. Added the RSH environment variable in order to be able to
1416 use a rsh replacement like ssh when doing remote backups (and
1417 bypass the security limitations of rcmd). Now you can do remote
1418 backups without being root (or making dump setuid root).
1419
1420 7. Modified again the way dumpdates works. For incremental dumps,
1421 we need to read dumpdates even if we are not using 'u' option.
1422 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
1423 this should work.
1424
1425 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
1426 ===================================================================
1427
1428 1. Removed the 'k' flag from the restore 'about' text if kerberos
1429 was not compiled in.
1430
1431 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
1432 to them (fsetflags takes a char*, setflags an open fd!).
1433
1434 3. (f)setflags is called only if the flags aren't empty. If the
1435 file is a special file, a warning is printed, because changing
1436 flags implies opening the device. Normally, a special file
1437 should not have any flag... (Debian bug #29775, patch provided
1438 by Abhijit Dasgupta <abhijit@ans.net>).
1439
1440 4. Made possible to dump a file system not mentioned in /etc/fstab.
1441 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
1442
1443 5. Changed the default behaviour to not create dumpdates
1444 unless 'u' option is specified. Removed the old "debian-patch"
1445 which provided the same thing. (Debian bug #38136, #33818).
1446
1447 6. Removed all those dump*announce, since they were getting old...
1448
1449 7. Added warning messages if dumpdates does not exist and
1450 when an inferior level dump does not exist (except for a level 0
1451 dump).
1452
1453 8. Debugged the glob calls in interactive mode: restore used a
1454 dirent struct which was different from the /usr/include/dirent.h
1455 one (this used to work, is it a glibc2 change?), so none of the
1456 compat glob (which used /usr/include/dirent.h) or the system glob
1457 worked. Restore use now the system dirent (and the system
1458 DT_* constants), which are compatible with BSD ones.
1459
1460 9. Added a configure flag (--with-dumpdatespath) to specify
1461 the location of dumpdates file. By default, it is
1462 /etc/dumpdates.
1463
1464 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1465 included the current date/version in man pages.
1466
1467 11. Corrected the estimation of remaining time when
1468 the operator doesn't change the tapes quickly enough. This
1469 was an old bug, I thought I corrected it, and discovered
1470 that in fact it was corrected in two different places, so
1471 the results canceled each other...
1472
1473 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1474 ===================================================================
1475
1476 1. Integrated multiple patches from RedHat, Debian and SuSE:
1477
1478 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1479 isn't specified.
1480 - several fixes in the man pages.
1481 - update the default tape device to /dev/st0.
1482 - many updates for Linux Alpha (byte ordering, size_t etc).
1483 - buffer overruns.
1484 - use environment variable for TMPDIR (instead of /tmp).
1485 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1486 - workaround egcs bug (RedHat bugs #4281 and #2989).
1487 - wire $(OPT) throughout Makefile's.
1488
1489 2. Upgrade the dump revision to 1, making possible to dump filesystems
1490 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1491
1492 3. Fix some compile warnings, prototype all functions.
1493
1494 4. Use glibc err/glob instead of internal compatibility
1495 routines (only if available).
1496
1497 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1498 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1499
1500 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1501 ======================================================================
1502
1503 1. Integrated the changes from FreeBSD-3.1-RELEASE
1504 (mostly bug fixes, buffer overruns, dump has now an "automatic
1505 tape length calculation" flag, dump/restore can use kerberos now
1506 (this is NOT tested), use environment variables for TAPE and
1507 RMT etc.).
1508
1509 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1510
1511 3. Corrected a bug in restore when using 'C' option with multi-volumes
1512 tapes (files splited accros two tapes give "size changed" errors
1513 when comparing).
1514
1515 4. Corrected the long standing bug when dumping multiple tapes.
1516 This works for me, needs further testing.
1517
1518 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1519 ====================================================================
1520
1521 1. Dump now runs correctly on kernels 2.1.x
1522 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1523
1524 Changes between versions 0.4b2 and 0.4b3
1525 ========================================
1526
1527 1. Use realpath() if available
1528
1529 2. Report statistics
1530
1531 Changes between versions 0.4b1 and 0.4b2
1532 ========================================
1533
1534 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1535 the patch)
1536
1537 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1538
1539 3. Fixed configure.in to correctly handle --enable-rmt
1540
1541 Changes between versions 0.3 and 0.4b1
1542 ======================================
1543
1544 1. Integrated the changes from 4.4BSD-Lite2
1545
1546 2. Integrated the patches from Debian and Red Hat
1547
1548 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1549
1550 4. Changed dump to use the Ext2fs library to get block addresses. This
1551 should solve the endianness problem on SparcLinux.
1552
1553 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1554 distribution's one) to use autoconf
1555
1556 6. Fixed a few minor bugs
1557
1558 Changes between versions 0.2e and 0.2f
1559 ======================================
1560
1561 1. Added the creation of named pipes (fifos) in restore.
1562
1563 2. Added the -N flag in the restore manual page.
1564
1565 3. Added the file kernel.patch which contains the llseek() optimization
1566 patch for 1.2.x kernels.
1567
1568 4. Fixed a bug in the restoration of symbolic links: owner and group were
1569 not restored.
1570
1571 5. Integrated some changes from FreeBSD 2.2.
1572
1573 6. Added a call to ftruncate() after restoring each file to restore
1574 correctly files ending by a hole.
1575
1576 Changes between versions 0.2d and 0.2e
1577 ======================================
1578
1579 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1580 with owner = root (0) and group = root/wheel/whatever (0).
1581
1582 Changes between versions 0.2c and 0.2d
1583 ======================================
1584
1585 1. Dump is now able to backup 2GB+ filesystems.
1586
1587 2. Dump and restore can now be linked as static binaries.
1588
1589 Changes between versions 0.2b and 0.2c
1590 ======================================
1591
1592 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1593
1594 Changes between versions 0.2a and 0.2b
1595 ======================================
1596
1597 1. Really fixed the bug that I should have corrected in 0.2a.
1598
1599 2. Enabled optimization again.
1600
1601 Changes between versions 0.2 and 0.2a
1602 =====================================
1603
1604 1. Disabled the optimization during compilation.
1605
1606 Changes between versions 0.1 and 0.2
1607 ====================================
1608
1609 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1610 the fs_type field (actually, I modified the file fstab.c to make it
1611 use the mntent functions).
1612
1613 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1614 made by dump should be readable by the BSD restore and backups made
1615 by the BSD dump should be readable by restore. Unfortunately, this
1616 means that the dump format has changed between version 0.1 and version
1617 0.2 :-(
1618
1619 3. Dump is now able to backup a subtree, it is no longer limited to whole
1620 filesystems like the BSD version.
1621
1622 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1623 than 2 GB.
1624
1625 Changes between versions 0.0 and 0.1
1626 ====================================
1627
1628 1. Now create links rdump and rrestore during the `make install' step.
1629
1630 2. Linux port specific bugs added to the manual pages
1631
1632 3. Incorrect estimation of the number of tapes blocks fixed when doing
1633 incremental backups.
1634
1635 4. Better ls-like format in restore in interactive mode.