Prepare for 0.4b38 release.
[dump.git] / CHANGES
1 $Id: CHANGES,v 1.263 2005/01/07 13:30:29 stelian Exp $
2
3 Changes between versions 0.4b37 and 0.4b38 (released January 7, 2005)
4 =====================================================================
5
6 1. Fix a couple of troff syntax bugs in the man pages.
7 Thanks to Eric Raymond <esr@thyrsus.com> for the patch.
8
9 2. Made restore use either libncurses or libtermcap, depending
10 on which one is available at configure time.
11
12 3. Fixed restore negative size display bug when comparing a
13 dump containing files over 2GB. Thanks to Steve Bonds
14 <sbonds@users.sourceforge.net> for the bug report.
15
16 4. Do not save directory entries to non-dumped inodes
17 (excluded from dump). This will eliminate the 'missing
18 file' warnings when doing 'restore -C'.
19
20 5. Fix dump crash when backuping a huge (2TB) filesystem,
21 due to a bogus calculation on the inode map size.
22 Thanks to Kevin B. Haines <K.B.Haines@rl.ac.uk> for
23 submitting the bug and testing the fix.
24
25 6. Fix a problem in restore where the final \0 in the symbolic
26 link names could have been lost, generating corrupt filenames.
27 Thanks to Kyle Wilson <kyle.wilson@amd.com> for reporting the
28 bug.
29
30 7. Implemented a hash list for the directory names in restore.
31 The linear list used before caused problems in interactive
32 restores when dealing with directories having thousands of
33 entries. Thanks to Brian Ristuccia <bristuccia@starentnetworks.com>
34 for reporting the bug.
35
36 8. Improved restore -C, this time including the directory
37 attributes into the comparision.
38
39 9. Made restore understand tapes containing EA/ACLs (which will
40 be dumped by the next version of dump). In this version
41 extended attributes on the tape are ignored, for full EA/ACL
42 support wait for the next version or try the experimental EA
43 patch.
44
45 Changes between versions 0.4b36 and 0.4b37 (released July 7, 2004)
46 ==================================================================
47
48 1. Added the --enable-staticz configure option which enables
49 dump and restore to be linked with static versions of
50 libz and libbz2 (and dynamic versions of all other libraries).
51 This will make Debian users happy, because libz and libbz2
52 were the only needed libraries living in /usr, all the
53 others live in /lib. In case of system emergency, it is
54 better not to have to rely on an extra filesystem.
55 Thanks to Bdale Garbee <bdale@gag.com> for the suggestion.
56
57 2. Fix compilation on (at least the Linux Debian port to) AMD64.
58 (<ext2fs/ext2_types.h> defines some types (__s64 and __u64)
59 that are also defined by <linux/types.h> (<asm/types.h>) and
60 they conflict).
61
62 3. Make dump's reading of the dumpdates file a bit more robust,
63 preventing dump from crashing when the dumpdates file has
64 been modified by hand.
65
66 4. Fixed some offset calculations in dump code which could
67 lead to "bread lseek errors" on large filesystems. Thanks
68 to Bruce Lowekamp <lowekamp@users.sourceforge.net> for
69 reporting this bug and debugging the issue.
70
71 5. Made dump use the blkid library when searching for devices
72 by label or uuid instead of dump's own routines.
73
74 6. Corrected a bug in dump where a wrong LABEL=... line in
75 /etc/fstab could prevent dump from dumping unrelated
76 filesystems. Thanks to Bruce Lowekamp
77 <lowekamp@users.sourceforge.net> for reporting the bug.
78
79 Changes between versions 0.4b35 and 0.4b36 (released April 21, 2004)
80 ====================================================================
81
82 1. Fixed dump compilation with old gcc versions. Thanks to
83 Mike Castle <dalgoda@users.sourceforge.net> for the patch.
84
85 2. Fixed some warnings (howmany, roundup, powerof2 redefined)
86 when compiling against a recent glibc version.
87
88 3. Fixed a bug in restore preventing the read of a dump tape
89 written with Solaris 7 ufsdump. Thanks to Patrick Higgins
90 <phiggins@transzap.com> for reporting the bug and providing
91 the test case.
92
93 4. Changed dump to enable the creation of volumes of different
94 sizes in a single run (make -B accept a list of values).
95 Patch contributed by Florian Zumbiehl <florz@gmx.de>.
96
97 5. Use the glibc provided minor() and major() macros instead
98 of our own bitmask implementation. This should be safe for
99 when the major/minor namespace will migrate to 32 bits.
100 Thanks to Zhang Jun <zhangjun@nanjing-fnst.com> for reporting
101 the bug.
102
103 6. Made explicit in the dump man page that dump will not create
104 a remote file, it will only write to an already existing one.
105
106 7. Another try at making size estimates better again.
107
108 8. Put back the inconditional running of the end-of-tape
109 script which was changed in 0.434 to be run only when -M
110 or multiple -f were NOT specified. Some users rely on this
111 feature even when it is combined with -M/-f.
112
113 9. Fixed restore when restoring huge backups (where rstdir...
114 temporary files are over 2GB). Thanks to
115 Raphael Jaffey <rjaffey@artic.edu> for reporting this,
116 debugging the issue and making the patch.
117
118 10. Made restore understand FreeBSD UFS2 tapes. Thanks to
119 David <vrtin@users.sourceforge.net> for submitting the bug
120 and providing a test case.
121
122 11. Made dump work with an arbitrary integer level (no more 10
123 levels only limitation). Thanks to Matthew
124 <msvincen@midway.uchicago.edu> for the patch.
125
126 Changes between versions 0.4b34 and 0.4b35 (released December 21, 2003)
127 =======================================================================
128
129 1. Added a note in the dump man page saying that the default
130 blocksize can be 32 if -d is used with a high density argument.
131 Thanks to Antonios Christofides <A.Christofides@itia.ntua.gr>
132 for the patch.
133
134 2. Fixed configure to correctly understand CPPFLAGS, CFLAGS,
135 LDFLAGS environment variables. Thanks to Arcady Genkin
136 <antipode@users.sourceforge.net> for reporting the bug.
137
138 3. Made -e/-E options of dump accept an unlimited number of inodes
139 to be excluded, rather than a hardcoded maximum. Thanks to
140 Dietrich Rothe <d-rothe@users.sourceforge.net> for the patch.
141
142 4. Updated the autoconf system to 2.50. Forced the -D_BSD_SOURCE
143 and -D_USE_BSD_SIGNAL defines in configure in order to solve
144 64bit build problems because quad_t is redifined with a
145 different signature. Thanks to Mike Harris <mharris@redhat.com>
146 for reporting this bug.
147
148 5. Made restore build on Solaris, making possible to
149 restore Linux's "enhanced" tapes. Thanks to Uwe Gohlke
150 <uwe@ugsoft.de> for the patch.
151
152 6. Made an extension in the dump tape format capable of saving
153 MacOSX specific inode extensions. Uwe Gohlke <uwe@ugsoft.de>
154 wrote the extension and contributed the restore code back
155 into this codebase. The same extension mechanism will be
156 used in the future to save ACLs...
157
158 7. Made rmt work correctly with regard to QFA and local/remote
159 files and tapes. The remote access will however work only
160 when the dump provided rmt version is used. If you want to
161 use another rmt server, please do not use the QFA feature.
162 Thanks to Uwe Gohlke <uwe@ugsoft.de> for the patch.
163
164 Changes between versions 0.4b33 and 0.4b34 (released April 18, 2003)
165 ====================================================================
166
167 1. Fixed the INSTALL file to reflect the actual install paths.
168 Thanks to David Ronis <ronis@ronispc.chem.mcgill.ca> for
169 reporting the bug.
170
171 2. Fixed the configure script to only check for headers presence
172 instead of trying to compile them. This should fix issues
173 with old build environments. Thanks to Kari Lempiainen
174 <kari@funky.pp.fi> for reporting the bug.
175
176 3. Fixed restore to correctly ignore sockets when comparing
177 a dump (as socket cannot be properly restored anyway). Thanks
178 to Gunther Reiszig <gunther@mit.edu> for reporting the bug.
179
180 4. Fixed restore to correctly access the archive file (-A argument)
181 even when using a remote tape. Thanks to Fabrice Bellet
182 <fabrice@bellet.info> for reporting the bug.
183
184 5. Fixed (again) handling of long (largefile) seeks in rmt.
185 Thanks to Fabrice Bellet <fabrice@bellet.info> for reporting
186 the bug.
187
188 6. Fixed restore corner case when dealing with large block sizes
189 dump is able to write now (-b 1024). Thanks to Fabrice Bellet
190 <fabrice@bellet.info> for reporting the bug.
191
192 7. Fixed a bug preventing dump to access a filesystem having
193 a label exactly 16 bytes in length. Thanks to <minduni@ti-edu.ch>
194 for reporting the bug.
195
196 8. Made dump store dump dates using explicit timezones, fixing a
197 problem with incremental dumps if the timezone is modified
198 between the dumps. Thanks to Philippe Troin <phil@fifi.org> for
199 the bug report and the patch.
200
201 9. Fixed a bug encountered when dumping individual files (not full
202 filesystems or directories) and dangling symbolic links happen
203 to be in the list of files. For as far as dump is concerned,
204 dangling symbolic links are allowed, and are dumped as is.
205 Thanks to Jin-su Ahn <jsahn@ee.snu.ac.kr> for reporting the
206 bug and providing the fix.
207
208 10. Fixed open and creation modes and permissions for QFA and
209 table-of-contents files in dump and restore. Thanks to
210 Philippe Troin <phil@fifi.org> for the patch.
211
212 11. Fixed the archive file descriptor handling enabling it to be 0.
213 This can happen in some cases when shell redirections are used.
214 Thanks to Philippe Troin <phil@fifi.org> for the patch.
215
216 12. Delayed the opening of archive file until after suid had been
217 dropped (fixing a possible security issue if dump is suid).
218 Thanks to Philippe Troin <phil@fifi.org> for the patch.
219
220 13. Fixed the 'S' command handling in the rmt client part.
221 Thanks to Philippe Troin <phil@fifi.org> for the patch.
222
223 14. Modified the end-of-tape script handling to print out statistics
224 (and stop the timer) before launching the eot script. Also, the eot
225 script does not get run anymore when using -M (which makes sense) or
226 when multiple tapes are listed on the command line
227 (-f tape0,tape1,tapen) (which also makes sense).
228 Thanks to Philippe Troin <phil@fifi.org> for the patch.
229
230 15. Relicensed dump/restore under the 'revised' BSD license, as per
231 ftp://ftp.cs.berkeley.edu/ucb/4bsd/README.Impt.License.Change.
232
233 16. Added LZO compression to dump. This new compression method has
234 the advantage of being super fast, thus not killing tape streaming
235 on slow machines. Thanks to Helmut Jarausch
236 <jarausch@igpm.rwth-aachen.de> for the patch and to
237 Markus Oberhumer <markus@oberhumer.com> for giving special permission
238 to include his miniLZO project (GPL licensed) in dump/restore.
239
240 17. Some small buffer overruns fixes in rmt. Thanks to Antonomasia
241 <ant@notatla.demon.co.uk> for reporting the bugs.
242
243 18. Added a special rmt version which can do encryption when writing
244 to tape. Read examples/encrypted_rmt/README for details on
245 how to enable and configure it. Thanks to Ken Lalonde
246 <ken@globalremit.com> for the patch.
247
248 19. Made dump work with 2.5 kernel end of tape early warning semantics.
249 Thanks to Kai Makisara <Kai.Makisara@kolumbus.fi> for the patch.
250
251 20. Fixed a bug which caused dump -w|-W not to work anymore, because
252 the fs_freq and fs_passno fields in /etc/mtab are always set
253 to 0 0. Thanks to Trent Piepho <xyzzy@speakeasy.org> for
254 reporting the bug.
255
256 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
257 =======================================================================
258
259 1. Added a note in the restore man page clarifying the question
260 'set the permissions on the current directory ?' asked by
261 restore at the end of treatment in -i and -x modes.
262
263 2. Fixed the endianess issues when reading compressed tapes.
264 Thanks to Dark Force <daq4th@users.sourceforge.net> for
265 reporting this bug and providing test cases.
266
267 3. Fixed the "ACL won't be dumped" warning message (which showed
268 an extra, unrelated error message). Thanks to Dragan Krnic
269 <dkrnic@lycos.com> for reporting this bug.
270
271 4. Made dump look first into /etc/mtab, then into /etc/fstab
272 when searching for filesystem paths. Also fixed some problems
273 caused by binding mounts or multiple block device mounts.
274 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
275 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
276 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
277 and #160305, Sourceforge bugs #618699 and #679832).
278
279 5. Made dump's -I option accept the value '0' meaning all the
280 read errors will be ignored. This can be useful when running
281 dump from unattended sessions (like cron jobs). Thanks to
282 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
283
284 6. Fixed the output of dump to indicate 'blocks' instead of
285 'tape blocks' in the various messages (blocks are always
286 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
287 made it clearly print the current blocksize at the start of
288 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
289 the suggestions.
290
291 7. Made rmt understand long (largefiles) seeks.
292
293 8. Fixed build with very old versions of libext2fs, where
294 EXT2_FT_* constants were undefined.
295
296 9. Made dump accept the dumpdates path on the command line
297 (-D file option) instead of using only the hardcoded one.
298 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
299 suggestion.
300
301 10. Enabled rmt, LFS, readline, QFA options by default in
302 ./configure. Updated the configure process (new versions
303 of config.guess, config.sub etc).
304
305 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
306 =======================================================================
307
308 1. Changed dump to use fcntl(F_SETLK) style locking instead
309 of flock() when locking the dumpdates file. With the old
310 locking scheme, a local user having read rights on the
311 dumpdates file could be able to do a Denial of Service attack
312 on dump. In order to lock the dumpdates file with the new
313 scheme, the user would need to have write access on the file.
314 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
315 reporting the bug (originally a bugtraq post).
316
317 2. Fixed interactive 'ls' which caused spurious errors warnings
318 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
319 <jorgen_ostling@users.sourceforge.net> for reporting this
320 bug.
321
322 3. Fixed dump's estimate when dealing with sparse inodes.
323
324 4. Modified dump to allow setting a blocksize bigger than 32kB
325 (raised the limit to 1024kB), because newer hardware needs
326 this for proper operation. Thanks to Dirk Traenapp
327 <dtraenapp@users.sourceforge.net> for reporting this.
328
329 5. Fixed a bug causing Dump to stop and report an error if an
330 inode number in the exclude file was followed by some amount
331 of whitespace. Thanks to Jeffrey Sofferin
332 <sofferin@users.sourceforge.net> for reporting this bug.
333
334 6. Fixed a bug which caused restore, in some particular cases,
335 to ask some 'scary' questions and leave a bunch of RSTTMP
336 directories behind when restoring incremental tapes. Thanks
337 to Philippe Troin <phil@fifi.org> for reporting this bug and
338 providing the test cases.
339
340 7. Changed the wording when inodes are excluded from dump:
341 replaced 'Added inode 7 to exclude list' with
342 'Excluding inode 7 (resize inode) from dump', as suggested
343 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
344
345 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
346 ===================================================================
347
348 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
349 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
350 <eros@lamel.bo.cnr.it> for reporting the bug.
351
352 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
353 ===================================================================
354
355 1. Made dump print out the ext2 logical block number in case of a read
356 error, which can be given as an argument to debugfs' ncheck command,
357 in order to find out the affected inode. Added note about this
358 usage in dump's man page.
359
360 2. Fixed a problem in restore when reading tapes written on big
361 endian machines with very old versions of dump. The patch was
362 contributed by George Helffrich <george@geology.bristol.ac.uk>.
363
364 3. Fixed the tape length calculation when using large tapes
365 and compression. Thanks to Georg Lippold
366 <g_lippold@sourceforge.net> for reporting the bug.
367
368 4. Added a new set of examples in dump_on_cd_2 directory, based
369 on dump_on_cd examples but somewhat enhanced, supporting DVD
370 media, and localized in english and german. Thanks to
371 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
372
373 5. Made dump save 32 bit UID/GID. Internally, this was achieved
374 by switching from the old BSD inode format to the new BSD
375 inode format, which means that the tape format was changed.
376 However, since all restore versions out there should
377 transparently support both inode formats, the change should
378 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
379 reporting the bug.
380
381 6. Fixed a lot of warnings in the code shown when compiling
382 with 'gcc -W'. Thanks to Matthias Andree
383 <matthias.andree@stud.uni-dortmund.de> for reporting this.
384
385 7. Fixed a small markup bug in the dump man page. Thanks to
386 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
387 patch.
388
389 8. Rewrote entirely the man pages using the tmac.an macro
390 package (Linux man page format) instead of the original BSD
391 format. They should be now cleaner and easier to modify.
392
393 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
394 ==================================================================
395
396 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
397 non Linux clients were misinterpreted. The description of the problem
398 (incompatible numbering in Linux mtio opcodes) is documented at
399 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
400 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
401 providing an excellent, cross-platform replacement for rmt in his
402 star package.
403
404 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
405 which caused restore to fail when the inode maps were not entirely
406 contained on the first volume (when using really small volumes or
407 when dumping a huge number of inodes).
408
409 3. Fixed a problem in dump, where files in subdirectories of directories
410 that have the nodump flag set (or being excluded with -e/-E) were
411 sometimes incorrectly dumped (instead of being excluded). The original
412 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
413 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
414 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
415 help testing the patch.
416
417 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
418 which makes one able to pipe the output of dump, by the net, to
419 a remote CD-burner server.
420
421 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
422 and use GNU's symbolic syntax over rmt instead of numerical values
423 to assure multiple platform compatibility.
424
425 6. Documented the -d option in restore.
426
427 7. Added a -v (verbose) mode to dump. For now it just prints the number
428 of the inode being dumped, but this could evolve in future versions
429 to include interesting debugging output.
430
431 8. Added a -o flag to restore, which automatically restores the current
432 directory permissions in -i or -x mode, without asking the operator
433 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
434 and Peter Samuel <peters@e-smith.com>.
435
436 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
437 ====================================================================
438
439 1. Fixed a bug in the QFA feature which could have caused
440 a bad tape positionning on restore (causing extra delays in
441 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
442
443 2. Added a small note in the dump man page specifying that
444 there should be no white space between the option letter and
445 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
446
447 3. Made restore work with older versions of the readline library,
448 by checking in configure for several extended symbols. Restore
449 can now be compiled with a readline as old as the 2.0 release
450 (though it may be a good idea to upgrade it to a more recent
451 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
452 reporting the build failures.
453
454 4. Fixed a performance problem with the QFA file creation in
455 dump, which made unnecessary seeks on the tape slowing down
456 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
457 reporting this issue.
458
459 5. Removed the inclusion of some kernel headers in the dump
460 source, which prevented the compile in some kernel/glibc
461 headers/architecture combination. Thanks to Bdale Garbee
462 <bdale@gag.com> for reporting the bug.
463
464 6. Added the appropriate error message when dump fails to
465 open the output file for writing. Thanks to Amith Varghese
466 <amithv@yahoo.com> for reporting this bug.
467
468 7. Made restore able to understand large Solaris ufsdump tapes
469 (containing inodes bigger than 4194304). Sun have introduced
470 an "extension" to the dump tape format when dealing with
471 those inodes, which was uncorrectly handled by Linux restore.
472 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
473 providing a test case.
474
475 8. Added the -m parameter to dump which optimises the output for
476 inodes having been changed but not modified since the last dump
477 ('changed' and 'modified' have the meaning defined in stat(2)).
478 For those inodes, dump will save only the metadata, instead of
479 saving the entire inode contents. Inodes which are either
480 directories or have been modified since the last dump are saved
481 in a regular way. Uses of this flag must be consistent, meaning
482 that either every dump in an incremental dump set have the flag,
483 or no one has it.
484
485 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
486 =======================================================================
487
488 1. Fixed behaviour of dump when exceeding resource limits
489 (SIGXFSZ treatment).
490
491 2. Added the -L flag to restore to allow the user to specify a
492 maximal allowed number of miscompares when using restore
493 with the -C option to check the backup.
494
495 3. Detailed the manual entry for the -N option of restore.
496
497 4. Added the -a flag to restore to make able doing unattended
498 restores in -i or -x mode (automatically walks through the
499 multiple dump volumes).
500
501 5. Extended the QFA mode to work with local files and/or
502 remote tapes and files. This way, restore can know in advance
503 the tape number and the offset for the inodes to extract and
504 can minimize the extraction time by seeking directly to the
505 good tape/offset.
506
507 6. Added the -A <archive> option to both dump and restore,
508 which makes dump to archive a dump table-of-contents in
509 the specified file to be used by restore to determine
510 whether a file is in the dump file that is being restored.
511 (the archive file syntax is also compatible with the
512 Solaris ufsdump generated one).
513
514 7. Small fix in restore making it able to read some (broken ?)
515 Solaris ufsdump tapes.
516
517 8. Fixed dump to correctly recognise the root filesystem when using
518 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
519 for reporting this bug.
520
521 9. Added the -P <file> option to restore to create a
522 Quick File Access file from an already made dump. Patch
523 contributed by Uwe Gohlke <uwe@ugsoft.de>.
524
525 10. Made restore compile and run on Solaris, making it a
526 possible replacement for the standard ufsrestore. Port was
527 contributed by Uwe Gohlke <uwe@ugsoft.de>.
528
529 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
530 =====================================================================
531
532 1. Added a set of backup scripts from Eugenio Diaz
533 <getnito@yahoo.com> in the examples section. It features
534 automatic (cron based) full and incremental dumping of
535 several filesystems on a separate filesystem.
536
537 2. Fixed a off-by-one miscalculation which disabled dumping
538 a one letter subdirectory. Thanks to Chris Siebenmann
539 <cks@utcc.utoronto.ca> for reporting the bug.
540
541 3. Fixed several restore bugs occuring when trying to
542 restore some missing files on the tape. Thanks to Chris
543 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
544
545 4. Fixed --with-ldopts configure argument passing, installing from
546 a separate object directory, makefile cleanups contributed
547 by <splite@purdue.edu>.
548
549 5. Fix a bug which could caused, in some conditions, the highest
550 number inode of a filesystem, to not be dumped. Many thanks
551 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
552 chase this bug.
553
554 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
555 =======================================================================
556
557 1. Added a mini howto from Patrick Walsh in the examples
558 directory.
559
560 2. Minor man pages syntax corrections. Thanks to
561 Chris Danis <screechco@home.com> for reporting the bugs.
562
563 3. Added a script from David B. Peterson <dave@toppledwagon.com>
564 to the examples section. It features dumping several
565 filesystems to a remote tape drive upon ssh.
566
567 4. Added a patch provided by Richard Jones <rich@annexia.org>
568 which allows BRADEMAX (number of read errors tolerated by
569 dump) to be adjusted using the -I option.
570
571 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
572 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
573 patch.
574
575 6. Add the -l option to restore to specify if, when doing a
576 remote restore, the file used to access the backup is a
577 regular file (the defaults being a tape drive). Restore needs
578 to know this information when reading a remote compressed
579 dump. Previously, this information was autodetected, but
580 the autodetection code fails (with ioctl: Inappropriate ioctl
581 for device) when using a non Linux remote box. Thanks to
582 many users and especially to Eros Albertazzi
583 <eros@lamel.bo.cnr.it> for reporting this.
584
585 7. Found a workaroung for the dump deadlock problem (3 childs
586 stuck in pause(), father in read()). The workaround seems
587 to work for me and several beta-testers. If it doesn't work
588 for you, please report back.
589
590 8. Updated the RPM spec file (BuildPrereq, URL etc).
591
592 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
593 ========================================================================
594
595 1. Fixed the permissions of a newly created QFA file by dump.
596
597 2. Cleaned up the source of dump (the external variables
598 definition was a complete mess, making possible to have
599 objects overlap).
600
601 3. Fixed restore to use the full tape volume path when doing
602 a compare (since it changes the working directory to the
603 filesystem being compared in the process).
604
605 4. Added the -q option to dump which makes dump abort
606 whenever it needs operator attention. It should be
607 easier to use dump in scripts with this option.
608
609 5. Detect the use of incompatible options to dump and
610 refuse them (like -a and -B options together).
611
612 6. Added bzip2 compression to dump/restore (use option -j level
613 to select it). Note that you will need a restore version
614 >= 0.4b24 in order to restore a bzip2 compressed dump.
615 The same warning as for the zlib compression applies:
616 the tape format of a bzip2 dump is not compatible with the
617 original BSD tape format.
618
619 7. Fixed a overflow problem in dump corrupting the dump when
620 very large files were encountered. Thanks to Vladimir Nadvornik
621 for the bug report.
622
623 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
624 kernel buffer/page cache before starting the dump, helping
625 a bit those who use dump on mounted filesystems. Thanks to
626 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
627 for this suggestion.
628
629 9. Updated the RPM spec file following the RedHat changes
630 (dynamically linked binaries now in /usr/sbin etc).
631
632 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
633 which enables restore to recognise multi volume compressed dumps
634 done on CD. Included his scripts for dump (which pipe the dump
635 output directly into cdrecord) and restore. There is now
636 possible to to backups to CD on the fly!
637
638 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
639 ===================================================================
640
641 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
642 Marcus Meissner (Caldera International Security Dept.).
643
644 2. Implement the Sun rmt extended protocol. Patch submitted
645 by Ian Gordon <iangordon@users.sourceforge.net>.
646
647 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
648 the linux kernel header. This ensures that dump always has the
649 latest version of this file. Patch submitted by
650 Andreas Dilger <adilger@turbolinux.com>.
651
652 4. Report any filesystem present in either /etc/fstab with a
653 positive passno or /etc/dumpdates in dump -w output.
654 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
655
656 5. Fixed the looping problem in dump introduced in the
657 previous version.
658
659 6. Changed the -B option of dump to limit the size of
660 _compressed_ output per volume if compression is on.
661 Patch contributed by Helmut Jarausch
662 <jarausch@igpm.rwth-aachen.de>. Note however that, since
663 it is impossible to predict the size of the compressed
664 data before doing the compression, the -B limit is a bit
665 conservative.
666
667 7. Fixed a bug in reading the operator typed file/tape path for
668 the next volume in restore.
669
670 8. Implemented a "-F script" option to restore which permits the
671 user to specify a script which will be launched at the
672 beginning of each tape, useful for automatic programming of
673 tape changers for example. See the restore man page for the
674 script parameters and return codes.
675
676 9. Small fix for the QFA routines provided by Uwe Gohlke
677 <uwe@ugsoft.de>, and some recommendations for QFA uses in
678 the man pages.
679
680 10. Fixed the multivolume restoring where making a mistake
681 to the 'Mount next tape' prompt caused several blocks to
682 be lost.
683
684 11. Enhanced the -e option of dump to take as a parameter a
685 comma separated list of inode numbers.
686
687 12. Added the -E option to dump which specify a file containing
688 inode numbers to exclude from the dump.
689
690 13. Fixed the compressed multi-volume dump + restore.
691
692 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
693 ==================================================================
694
695 1. Made dump report the number of blocks dumped per volume.
696 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
697
698 2. Fix a bug in dump where waiting too much at the 'change volumes'
699 question caused the volume to be rewritten. Thanks to
700 Graham Reed <greed@users.sourceforge.net> for reporting the
701 bug and providing a patch.
702
703 3. Added a compression option to dump, contributed by Jerry
704 Peters <gapeters@worldnet.att.net>.
705
706 WARNING: the compression works only on files/pipes and
707 drives supporting variable block size.
708
709 WARNING2: tapes created using the compression feature are
710 incompatible with the standard dump tape format, and a
711 version of restore >= 0.4b22 will be needed for extraction.
712
713 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
714 architectures. Thanks to Paul Slootman <paul@debian.org> for
715 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
716 it upstream.
717
718 5. Many cleanups (CPP defines, const char warnings, check of
719 ext2fs COMPAT flags, time_t cleanups, added libext2 version
720 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
721
722 6. Made --prefix option work in configure. All the install path
723 are now based on the configure parameters.
724
725 7. Added the Quick File Access mode in dump/restore, contributed
726 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
727 a file tape position for each inode, and this file is used by
728 restore (if called with parameter Q and the filename)
729 to directly position the tape at the file restore is currently
730 working on. This saves hours when restoring single files from
731 large backups, saves the tapes and the drive's head. Use
732 --enable-qfa option of configure to compile in the QFA support.
733
734 8. Added the possibility to dump several files and directories
735 in a single invocation of dump. Thanks to Uwe Gohlke
736 <uwe@ugsoft.de> for implementing this option.
737
738 9. Fixed the dumping and restoring of files with holes
739 (files ending with holes didn't get dumped or restored
740 correctly).
741
742 10. Fixed a socket descriptor leak in dump, which leaved opened
743 3 file descriptors per dump process (and there is one dump
744 process per tape).
745
746 11. Fixed dump large file system support, by explicit use of
747 open64/lseek64/etc functions (explicit use needed because
748 e2fsprogs libraries don't behave well when compiled with
749 FILE_OFFSET_BITS=64).
750
751 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
752 ======================================================================
753
754 1. Fixed some bugs in the dump -w|-W logic introduced by
755 the previous version. Thanks to Andreas Dilger
756 <adilger@turbolinux.com> for his help on this one.
757
758 2. Fixed again a compilation problem when using old e2fs
759 headers (filesystem label related). Thanks to many users
760 who reported this stupid error.
761
762 3. Fixed a build problem on old lib5 based systems dealing with
763 _PATH_MNTTAB being undefined. Thanks to John Adams
764 <johna@onevista.com> for reporting the bug.
765
766 4. Improved the error detection in restore when creating
767 the temporary files in TMPDIR. Now restore will corectly
768 report a 'no space left on device' error instead of
769 strange errors which could imply an archive corruption.
770 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
771 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
772
773 5. Added the throughput information in dump's progression
774 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
775 for the patch.
776
777 6. Use libext2fs's inode scanning routines, which are particularly
778 robust when dealing with errors (especially when having some
779 bad blocks on the filesystem). Many thanks to Colin
780 <colin@horizon.com> for reporting the bug, investigating
781 the issues, finding a workaround, writing the patch and
782 fully testing it... (of course, if this patch will break
783 anything, he is to blame for :-)).
784
785 7. Made dump and restore LFS aware. Dump can dump a filesystem
786 containing large files, generate a large file on output and
787 restore can restore them. This feature is not enabled by
788 default, use --enable-largefile option of configure to enable
789 it (you need to have a LFS aware glibc though). Thanks to
790 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
791 and to Theodore T'so <tytso@valinux.com> for his always
792 useful thoughts.
793
794 8. Made dump ask upon a tape write error if it should rewrite
795 the current volume (assume this is a bad tape) or if it should
796 assume an end-of-tape condition (useful for tape drives which
797 behaves badly at the end of the tape). Thanks to Andreas
798 Dilger <adilger@turbolinux.com> for the suggestion.
799
800 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
801 =======================================================================
802
803 1. Fixed a small compilation problem due to a change
804 in the definintion of the struct sigaction in
805 glibc 2.0 / libc5. Thanks to Gunther Schlegel
806 <schlegel@riege.de> for reporting the bug and to
807 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
808 a fix.
809
810 2. Modified the label and uuid analysis in order to be
811 self-contained (independant of kernel/libc headers). This
812 should fix the compile with older kernel/libc headers and
813 will preserve the functionality. Thanks to Bernhard Erdmann
814 <bernhard.erdmann@gmx.de> for reporting the bug.
815
816 3. The 'exclude inode' option, if applied to a directory
817 inode, excludes now correctly all the leaves of this
818 directory. Thanks to John R. Dennison
819 <gerdesas@users.sourceforge.net> for reporting the bug.
820
821 4. Fixed the '-e' option to disable the possibility
822 to exclude the root inode (which causes the dumps to
823 be unreadable by restore). Prevented array overflow
824 when multiple -e options are used.
825
826 5. Fixed dump to correctly interpret a filesystem argument
827 which is a mountpoint and it is not an absolute path
828 (as specified in the fstab). Thanks to Bernhard R. Erdmann
829 <be@berdmann.de> for reporting the bug.
830
831 6. Made dump able to backup files larger than 2 GB. Note that
832 dump still doesn't cope with files larger than 4 GB.
833
834 7. Restore the real uid and gid before invoking an external
835 RSH program (big hole when dump or restore is suid root!).
836
837 8. Get the values for the default tape device and the location
838 of fstab file from the system headers. Thanks to
839 Andreas Dilger <adilger@turbolinux.com> for the patch.
840
841 9. Made dump -w|-W report all recognized filesystems
842 present in either /etc/fstab or /etc/dumpdates, and present
843 the list in the same order as in fstab file. Thanks
844 to Andreas Dilger <adilger@turbolinux.com> for the patch.
845
846 10. Made dump's -a (automatic end of tape detection) the
847 default. Specifying one of -B, -c, -d or -s options will
848 override the EOM detection. Thanks to Andreas Dilger
849 <adilger@turbolinux.com> for the patch.
850
851 11. Save the ext2 filesystem label into the dump volume label.
852 Specifying a volume label on the command line (-L option)
853 will override this feature. Thanks to Andreas Dilger
854 <adilger@turbolinux.com> for the patch.
855
856 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
857 =====================================================================
858
859 1. Fixed the signal handling in dump (which I broke in 0.4b17)
860 which was causing several strange dump failures (dump
861 hanged or segmentation faults etc).
862
863 2. Specified the default blocksize in the dump man page.
864
865 3. Changed two info messages of restore to be written on stdout
866 instead of stderr in order to leave stderr only for errors
867 or warnings. Thanks to Stephen Atwell
868 <satwell@urbana.css.mot.com> for the suggestion.
869
870 4. Corrected an off by one calculation which prevented
871 dumping a directory having a 1 character name length.
872 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
873 for reporting the bug.
874
875 5. Reinforce dump in order to allow only level 0 and no
876 -u option when dumping a subdirectory, and document
877 this in the man page. Thanks to Bernhard Erdmann
878 <bernhard.erdmann@gmx.de> for reporting the bug.
879
880 6. Fixed a small harmless bug in restore which caused
881 in some conditions hard links to be restored several
882 times (therefore generation some warning messages).
883 Thanks to Kenneth Porter <shiva@well.com> for
884 reporting the bug.
885
886 7. Updated the RPM spec file to the latest RedHat version,
887 providing FHS packaging and other cosmetic changes.
888 You will need RPM version >= 3.0.5 in order to build the RPM.
889
890 8. Updated the configure script to check for libtermcap
891 before checking for libreadline, because we need this
892 library in order to compile the readline support.
893
894 9. Made dump understand the LABEL= and UUID= notation
895 both in /etc/fstab and on the command line. Note that
896 you will need the /proc filesystem in order to use
897 these notations. Thanks to Erik Troan <ewt@redhat.com>
898 for providing the patch.
899
900 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
901 ===================================================================
902
903 1. Fixed a potential buffer overflow in restore. Thanks
904 to Stan Bubrouski <satan@fastdial.net> for reporting
905 the bug.
906
907 2. Fixed a readline-related bug which prevented
908 'cat DUMPFILE | restore -i -f -' from working. Thanks
909 to Charles Karney <karney@users.sourceforge.net>
910 for the bug report.
911
912 3. Changed a few "panic" into "exit", causing restore to
913 be more stable against some attacks (like the last one
914 reported on Bugtraq, although the last version of restore
915 was not vulnerable - just dumped core). Thanks to
916 Andreas Hasenack <andreas@conectiva.com.br> for reporting
917 the bugs to me.
918
919 4. Removed the suid-root bit on dump and restore in the
920 default build (and generated RPMs). It should be safer
921 now. Users who need the suid-root capabilities in order
922 to make network backups should read first the man page
923 of dump and enable the suid bit manually.
924
925 5. Added -ltermcap to the compile parameters for restore
926 when using readline support, in order to make the compile
927 process work on some systems (SuSE ?). Thanks to
928 Patrik Schindler <poc@pocnet.net> for reporting the bug.
929
930 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
931 ====================================================================
932
933 1. The -F script is called now *only* at the end of a tape,
934 not at the end of the dump. Thanks to Christian Haul
935 <haul@informatik.tu-darmstadt.de> for the bug report.
936
937 Normally, the device name and the current volume number
938 are passed on the command line of the script. However,
939 if you want the old style script (with no arguments
940 passed on the command line), you can enable it in
941 configure with the --enable-oldstylefscript.
942
943 2. Use posix signal handling to preserve dump functionality
944 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
945 the patch.
946
947 3. Made the exit code of restore in case of a 'C'ompare
948 command reflect the result of the compare. An exit status
949 of 0 means the dump archive is correct, 1 means tape errors,
950 2 means that some files were modified. Thanks to Kenneth Porter
951 <shiva@well.com> for the suggestion.
952
953 4. Made (finally) quotes work in interactive restore.
954
955 5. Small fixes in order to allow dump to compile with a
956 really old e2fsprogs version. Thanks to Ian Zimmerman
957 <itz@speakeasy.org> for the bug report.
958
959 6. Add GNU readline capabilities to interactive restore.
960 Use configure's --enable-readline flag to enable this feature.
961 Thanks to Patrik Schindler <poc@pocnet.net> for the
962 suggestion.
963
964 7. Do the compare on the fly in restore 'C' mode (this will
965 allow not to exhaust the available /tmp space when
966 ccmparing large files). Thanks to Kenneth Porter
967 <shiva@well.com> for the suggestion.
968
969 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
970 ====================================================================
971
972 1. Fixed some several duplicate 'DUMP: DUMP:' in the
973 output of dump.
974
975 2. Corrected the estimation of blocks to dump. Note that
976 this is still wrong for large directory inodes, due
977 to the size of a BSD directory entry on the tape
978 (estimation is based on the size of the ext2 directory,
979 which is smaller because it doesn't contain the null
980 character at the end).
981
982 3. Made dump report the total number of MB written to
983 tape at the end of run. Thanks to W. Reilly Cooley
984 <wcooley@nakedape.cc> for the patch.
985
986 4. Added the -X option to restore giving the possibility
987 to read the names of the files to be extracted/listed
988 from a text file (in addition of the command line).
989 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
990 patch.
991
992 5. Added the device name and the current volume number
993 as arguments to the end of tape script (dump -F option).
994
995 6. Made the multi-volume dump work again (after having
996 broken it in 0.4b15).
997
998 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
999 ===================================================================
1000
1001 1. Added a prompt command in interactive restore mode. Thanks
1002 to Andreas Dilger <adilger@home.com> for the patch.
1003
1004 2. Fixed a buffer overflow problem in dump (caused by
1005 not checking the size of the filesystem parameter).
1006 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
1007 reporting this on Bugtraq (and to several dump users
1008 who forwarded me his mail).
1009
1010 3. Added the '-F script' option to dump in order to
1011 launch a script at the end of each tape (to be used
1012 with a tape changer, or to notify the sysadmin by
1013 pager etc.).
1014
1015 4. Fixed a bug in restore compare code caused by the changes
1016 I made in 0.4b14.
1017
1018 5. Fixed the treatment of options using the old BSD syntax
1019 in both dump and restore.
1020
1021 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
1022 =======================================================================
1023
1024 1. Fixed a bug in dump which may have caused invalid deleted
1025 directories to be dumped out if they were deleted after the
1026 mapping pass. This could occure on active filesystem and lead
1027 to heap corruption (causing dump malfunction in many possible ways).
1028 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
1029 bug and submitting the patch.
1030
1031 2. Fixed the handling of the filesystem parameter in dump. A
1032 '/mnt/disk' parameter caused the disk contents to be dumped,
1033 but a '/mnt/disk/' parameter caused the mountpoint directory
1034 to be dumped (generally an empty directory).
1035
1036 3. Improved the output of dump in order to tell which directory
1037 it is currently dumping (when dumping a subtree).
1038
1039 4. Added the '-e' exclude inode option to dump. Thanks to
1040 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
1041
1042 5. Added a REPORTING-BUGS file in order to provide a guide
1043 on how to correctly report a bug in dump/restore.
1044
1045 6. Fixed a restore problem when restoring a hard link to an inode
1046 having the immutable or append-only attribute set. Thanks to
1047 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
1048
1049 7. Fixed a compatibility problem between dumps made on little
1050 endian machines (the format was correct) and big endian
1051 machines (the format was incorrect). This fix break the
1052 compatibility with the older dumps made on big endian
1053 machines (sparc, m86k, ppc etc). For the first time in
1054 linux dump's history (I believe), the dumps made by *BSD,
1055 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
1056 so interchangeable. Thanks to Rob Cermak
1057 <cermak@ahab.rutgers.edu> for submitting the bug and
1058 helping me test the fix.
1059
1060 8. Fixed the way dump reports the remaining percent/time, if
1061 the number of blocks actually dumped exceeds the estimated
1062 number of blocks. Thanks to Jean-Paul van der Jagt
1063 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
1064
1065 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
1066 ======================================================================
1067
1068 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
1069 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
1070
1071 2. Removed the requirement to build the RPM as root from the
1072 spec file. Thanks to Christian Weisgerber
1073 <naddy@mips.rhein-neckar.de> for submitting this.
1074
1075 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
1076 causing dump to block when dumping really huge filesystems.
1077 Many thanks to Patrik Schindler <poc@pocnet.net> for
1078 helping me find this bug.
1079
1080 4. Fixed the treatment for an interrupt signal when dump access
1081 the remote tape through RSH. Thanks to Christian Weisgerber
1082 <naddy@mips.rhein-neckar.de> for providing the patch.
1083
1084 5. Fixed a bug which was causing dump/restore to display
1085 garbage characters instead of the remote host name.
1086
1087 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
1088 =====================================================================
1089
1090 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
1091 <kukuk@suse.de> for submitting the patch.
1092
1093 2. Fix for the exit code when using the size estimate option of
1094 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
1095 submitting the patch.
1096
1097 3. Handle EINTR in atomical reads/writes in dump, which was causing
1098 dump to fail on some systems. Thanks to Eric Jergensen
1099 <eric@dvns.com> for reporting the bug and submitting the patch.
1100
1101 4. Handle more than 16 characters for the device names in dumpdates.
1102 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
1103 tracking down the problem and proposing the solution.
1104
1105 5. Fixed a bug in dump which prevented the creation of the
1106 dumpdates file when doing a 0-level dump without already
1107 having a dumpdates file. Thanks to Patrik Schindler
1108 <poc@pocnet.net> for reporting the bug.
1109
1110 6. Changed the way dump 'S' flag reports the size estimate
1111 from number of blocks into bytes (making it compatible
1112 with the Solaris version, and simplifying things for
1113 amanda users). Thanks to Jason L Tibbitts III
1114 <tibbs@math.uh.edu> for reporting the bug.
1115
1116 7. Fixed a compatibility problem in linux/alpha dump tape format.
1117 Now the linux/alpha dump are (again) compatible with the
1118 other dump formats. But this breaks compatibility with
1119 older dumps made on alpha. Thanks to Mike Tibor
1120 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
1121
1122 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
1123 ======================================================================
1124
1125 1. Added a '--enable-kerberos' to configure.
1126
1127 2. Added a 'S' option to dump which determines the amount of space
1128 that is needed to perform the dump without actually doing it, similar
1129 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
1130 <cermak@ahab.rutgers.edu>.
1131
1132 3. Added a 'M' multi-volume option to dump and restore which enables
1133 dumping to multiple files (useful when dumping to an ext2
1134 partition to make several dump files in order to bypass the 2GB
1135 file size limitation). The 'f' argument is treated as a prefix and
1136 the output files will be named <prefix>001, <prefix>002 etc. With
1137 the 'M' flag, restore automatically selects the right file without
1138 asking to enter a new tape each time.
1139
1140 4. Fixed a memory leak which was causing dump to grow very big
1141 (270MB when dumping a 10GB filesystem...). Thanks to Jason
1142 Fearon <jasonf@netrider.org.au> for reporting the bug.
1143
1144 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
1145 ======================================================================
1146
1147 1. Make configure test if the system glob routines support
1148 extended syntax (ALTDIRFUNC). If not, use the internal glob
1149 routines instead of system ones. Thanks to Bernhard Sadlowski
1150 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
1151 and helping me resolve this and other minor libc5 compiling
1152 glitches.
1153
1154 2. Fix a problem when dumping a ext2fs with the 'filetype'
1155 feature enabled. Thanks to Patrick J. LoPresti
1156 <patl@cag.lcs.mit.edu> for reporting the bug and to
1157 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
1158
1159 3. Made the nodump flag work on directories. A directory which
1160 has the nodump flag gets never dumped, regardless of its
1161 contents.
1162
1163 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
1164 which allows dump on an active ext3 filesystem. However, this
1165 is a "quick and dirty" patch which enables backup of an ext3
1166 filesystem through the ext2 compatibility (by ignoring the
1167 NEEDS_RECOVERY bit). The journal file is not recognized and
1168 it is dumped (it should not).
1169
1170 5. Test the superblock compatibility flags when dumping, in order
1171 to be sure that we know how to deal with specific features.
1172
1173 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
1174 ====================================================================
1175
1176 1. Use lchown instead of chown, fixing a possible security problem
1177 when restoring symlinks (a malicious user could use this
1178 to deliberately corrupt the ownership of important system files).
1179 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
1180 this and providing the patch.
1181
1182 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
1183 ====================================================================
1184
1185 1. Put dump sources under CVS, added Id tags in all files so
1186 one can use 'ident' on binary files.
1187
1188 2. Added the dump/restore version in the usage text so one can
1189 easily verify the version he is using.
1190
1191 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
1192 a va_start/va_end problem on linux-ppc (always call va_start
1193 va_end in pairs each time we use a vararg function).
1194
1195 4. Added again the DT_* constants because old libc does not
1196 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
1197 for submitting the bug report.
1198
1199 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
1200 this should enable dumping big (huge) filesystems.
1201
1202 6. Added the RSH environment variable in order to be able to
1203 use a rsh replacement like ssh when doing remote backups (and
1204 bypass the security limitations of rcmd). Now you can do remote
1205 backups without being root (or making dump setuid root).
1206
1207 7. Modified again the way dumpdates works. For incremental dumps,
1208 we need to read dumpdates even if we are not using 'u' option.
1209 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
1210 this should work.
1211
1212 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
1213 ===================================================================
1214
1215 1. Removed the 'k' flag from the restore 'about' text if kerberos
1216 was not compiled in.
1217
1218 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
1219 to them (fsetflags takes a char*, setflags an open fd!).
1220
1221 3. (f)setflags is called only if the flags aren't empty. If the
1222 file is a special file, a warning is printed, because changing
1223 flags implies opening the device. Normally, a special file
1224 should not have any flag... (Debian bug #29775, patch provided
1225 by Abhijit Dasgupta <abhijit@ans.net>).
1226
1227 4. Made possible to dump a file system not mentioned in /etc/fstab.
1228 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
1229
1230 5. Changed the default behaviour to not create dumpdates
1231 unless 'u' option is specified. Removed the old "debian-patch"
1232 which provided the same thing. (Debian bug #38136, #33818).
1233
1234 6. Removed all those dump*announce, since they were getting old...
1235
1236 7. Added warning messages if dumpdates does not exist and
1237 when an inferior level dump does not exist (except for a level 0
1238 dump).
1239
1240 8. Debugged the glob calls in interactive mode: restore used a
1241 dirent struct which was different from the /usr/include/dirent.h
1242 one (this used to work, is it a glibc2 change?), so none of the
1243 compat glob (which used /usr/include/dirent.h) or the system glob
1244 worked. Restore use now the system dirent (and the system
1245 DT_* constants), which are compatible with BSD ones.
1246
1247 9. Added a configure flag (--with-dumpdatespath) to specify
1248 the location of dumpdates file. By default, it is
1249 /etc/dumpdates.
1250
1251 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1252 included the current date/version in man pages.
1253
1254 11. Corrected the estimation of remaining time when
1255 the operator doesn't change the tapes quickly enough. This
1256 was an old bug, I thought I corrected it, and discovered
1257 that in fact it was corrected in two different places, so
1258 the results canceled each other...
1259
1260 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1261 ===================================================================
1262
1263 1. Integrated multiple patches from RedHat, Debian and SuSE:
1264
1265 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1266 isn't specified.
1267 - several fixes in the man pages.
1268 - update the default tape device to /dev/st0.
1269 - many updates for Linux Alpha (byte ordering, size_t etc).
1270 - buffer overruns.
1271 - use environment variable for TMPDIR (instead of /tmp).
1272 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1273 - workaround egcs bug (RedHat bugs #4281 and #2989).
1274 - wire $(OPT) throughout Makefile's.
1275
1276 2. Upgrade the dump revision to 1, making possible to dump filesystems
1277 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1278
1279 3. Fix some compile warnings, prototype all functions.
1280
1281 4. Use glibc err/glob instead of internal compatibility
1282 routines (only if available).
1283
1284 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1285 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1286
1287 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1288 ======================================================================
1289
1290 1. Integrated the changes from FreeBSD-3.1-RELEASE
1291 (mostly bug fixes, buffer overruns, dump has now an "automatic
1292 tape length calculation" flag, dump/restore can use kerberos now
1293 (this is NOT tested), use environment variables for TAPE and
1294 RMT etc.).
1295
1296 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1297
1298 3. Corrected a bug in restore when using 'C' option with multi-volumes
1299 tapes (files splited accros two tapes give "size changed" errors
1300 when comparing).
1301
1302 4. Corrected the long standing bug when dumping multiple tapes.
1303 This works for me, needs further testing.
1304
1305 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1306 ====================================================================
1307
1308 1. Dump now runs correctly on kernels 2.1.x
1309 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1310
1311 Changes between versions 0.4b2 and 0.4b3
1312 ========================================
1313
1314 1. Use realpath() if available
1315
1316 2. Report statistics
1317
1318 Changes between versions 0.4b1 and 0.4b2
1319 ========================================
1320
1321 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1322 the patch)
1323
1324 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1325
1326 3. Fixed configure.in to correctly handle --enable-rmt
1327
1328 Changes between versions 0.3 and 0.4b1
1329 ======================================
1330
1331 1. Integrated the changes from 4.4BSD-Lite2
1332
1333 2. Integrated the patches from Debian and Red Hat
1334
1335 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1336
1337 4. Changed dump to use the Ext2fs library to get block addresses. This
1338 should solve the endianness problem on SparcLinux.
1339
1340 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1341 distribution's one) to use autoconf
1342
1343 6. Fixed a few minor bugs
1344
1345 Changes between versions 0.2e and 0.2f
1346 ======================================
1347
1348 1. Added the creation of named pipes (fifos) in restore.
1349
1350 2. Added the -N flag in the restore manual page.
1351
1352 3. Added the file kernel.patch which contains the llseek() optimization
1353 patch for 1.2.x kernels.
1354
1355 4. Fixed a bug in the restoration of symbolic links: owner and group were
1356 not restored.
1357
1358 5. Integrated some changes from FreeBSD 2.2.
1359
1360 6. Added a call to ftruncate() after restoring each file to restore
1361 correctly files ending by a hole.
1362
1363 Changes between versions 0.2d and 0.2e
1364 ======================================
1365
1366 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1367 with owner = root (0) and group = root/wheel/whatever (0).
1368
1369 Changes between versions 0.2c and 0.2d
1370 ======================================
1371
1372 1. Dump is now able to backup 2GB+ filesystems.
1373
1374 2. Dump and restore can now be linked as static binaries.
1375
1376 Changes between versions 0.2b and 0.2c
1377 ======================================
1378
1379 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1380
1381 Changes between versions 0.2a and 0.2b
1382 ======================================
1383
1384 1. Really fixed the bug that I should have corrected in 0.2a.
1385
1386 2. Enabled optimization again.
1387
1388 Changes between versions 0.2 and 0.2a
1389 =====================================
1390
1391 1. Disabled the optimization during compilation.
1392
1393 Changes between versions 0.1 and 0.2
1394 ====================================
1395
1396 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1397 the fs_type field (actually, I modified the file fstab.c to make it
1398 use the mntent functions).
1399
1400 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1401 made by dump should be readable by the BSD restore and backups made
1402 by the BSD dump should be readable by restore. Unfortunately, this
1403 means that the dump format has changed between version 0.1 and version
1404 0.2 :-(
1405
1406 3. Dump is now able to backup a subtree, it is no longer limited to whole
1407 filesystems like the BSD version.
1408
1409 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1410 than 2 GB.
1411
1412 Changes between versions 0.0 and 0.1
1413 ====================================
1414
1415 1. Now create links rdump and rrestore during the `make install' step.
1416
1417 2. Linux port specific bugs added to the manual pages
1418
1419 3. Incorrect estimation of the number of tapes blocks fixed when doing
1420 incremental backups.
1421
1422 4. Better ls-like format in restore in interactive mode.