Add ext4 filesystem support.
[dump.git] / CHANGES
1 $Id: CHANGES,v 1.301 2009/06/18 10:00:38 stelian Exp $
2
3 Changes between versions 0.4b41 and 0.4b42 (released xxxxxxxxxxxxxxx)
4 =====================================================================
5
6 1. Fix printout of the 'Connection to' message in dump/restore.
7 Thanks to Dale Wiles <dwiles@users.sourceforge.net> for the
8 bug report and its fix.
9
10 2. Fix dump -w/-W output for filesystems having the last backup
11 level equal to 0. Thanks to Pascal Bouchareine
12 <kalou@users.sourceforge.net> for reporting the bug.
13
14 3. Enable restore to handle restoring onto a different SELinux
15 policy from the dump. Thanks to Tony Nelson
16 <tony_nelson@users.sourceforge.net> for the patch.
17 (see http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=189845
18 for details)
19
20 4. Fix EA set failures when restoring immutable files. Thanks
21 to Andrew Kroeger for the patch.
22
23 5. Fix "mode file too large" errors when restoring huge backups.
24 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
25 report.
26
27 6. Add the ability to link against libtinfo library if present,
28 instead of libtermcap or libncurses libraries. Thanks to
29 Adam Tkac <atkac@redhat.com> for the patch.
30
31 7. When comparing a backup, do not consider to be an error the fact
32 that we encounter files created while dump was in progress.
33 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
34 report.
35
36 8. Force '-a' to be the default value when running 'restore -C'.
37 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
38 report.
39
40 9. Many improvements to the 'cron_dump_to_disk' example.
41 Thanks to Aaron S. Hawley <ashawley@users.sourceforge.net> for
42 the patch.
43
44 10. Fix the QFA generation when extended attributes are backuped
45 along with the inodes. Prior to this fix, the entries for
46 some inodes may get corrupted (duplicated entries or incorrect
47 ones), making the QFA file unusable for repositionning in
48 restore. Many thanks to Kenneth Porter <shiva@sewingwitch.com>
49 for the bug report and the testing of the fix.
50
51 11. Fix EA restoration in debug mode (Sourceforge bug #1986216).
52
53 12. Stefan Auracher <star7@onlinehome.de> noticed that the Tower
54 of Hanoi backup strategy documented in the man page wasn't
55 actually based on the Tower of Hanoi algorithm. The man page
56 was updated accordingly.
57
58 13. Fix missing level information from dump output when using
59 the default level (RedHat bug #493635). Thanks to Adam Tkac
60 <atkac@redhat.com> for the patch.
61
62 14. Fix the dump man page to reflect the fact that the default
63 dump level is 0 not 9 (RedHat bug #356121). Thanks to Adam Tkac
64 <atkac@redhat.com> for the patch.
65
66 15. Fixed a few spelling errors in the man pages (RedHat bug #489853).
67 Thanks to Adam Tkac <atkac@redhat.com> for the patch.
68
69 16. Fixed restore man page and program usage which incorrectly stated
70 that -P and -A were both allowed in the same invocation (RedHat
71 bug #490627). Thanks to Adam Tkac <atkac@redhat.com> for the patch.
72
73 17. Use sys_clone under Linux to share I/O contexts between dump
74 processes, thus drastically increasing the performance of dump
75 under CFQ (which is the default I/O scheduler used in a number of
76 distributions). Thanks to Jeff Moyer <jmoyer@redhat.com> for the
77 patch.
78
79 18. Add (preliminary) ext4 support - thanks to libext2fs which does
80 all the job for us. Thanks to Gertjan van Wingerde
81 <gwingerde@gmail.com> for the patch.
82
83 Changes between versions 0.4b40 and 0.4b41 (released January 2, 2006)
84 =====================================================================
85
86 1. Fix restore of dumped Access Control Lists. The previous
87 code for EA works fine, but ACLs needed conversion from the
88 ext2/3 disk format to posix_acl format before restoring.
89
90 2. Fix some issues with restoration of EA on big endian
91 platforms.
92
93 3. Fix restore when the symtab is over 2GB in size.
94
95 4. Made the directory hash indexing an optional feature,
96 accessible by the '-H' option of restore, and disable it
97 by default.
98
99 5. Fixed dump to not include extended attributes information
100 in the toc (archive) file which confused restore -t.
101
102 Changes between versions 0.4b39 and 0.4b40 (released May 2, 2005)
103 =================================================================
104
105 1. Changed restore to emit warnings (instead of emitting a fatal
106 error) if a file (or a directory) is unavailable for a
107 comparision (if the user doesn't have the necessary permissions
108 to access it for example). Thanks to Kenneth Porter
109 <shiva@sewingwitch.com> for the bug report.
110
111 2. Re-done the 'do not save directory entries to non-dumped inodes
112 (excluded from dump)' feature. The previous implementation
113 worked well for excluded directories but not for regular files.
114 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
115 report.
116
117 3. Fixed a bug in dump where the tape size was miscalculated when
118 the user used -d/-s to specify the tape characteristics. Thanks
119 to Philip Goisman <goisman@physics.arizona.edu> for reporting
120 the bug.
121
122 4. Fixed another bug introduced in restore with the hashtree
123 implementation. This one caused restore to stop saying
124 "removenode: non-empty directory" in some cases.
125
126 5. Added support for dumping and restoring ext2/3 extended
127 attributes (EA), like the access control lists (ACL) or
128 the security labels used by SELinux.
129
130 Changes between versions 0.4b38 and 0.4b39 (released January 21, 2005)
131 ======================================================================
132
133 1. The newly added dump_on_cd_3 example was buggy, replace it
134 with an updated version from Andrew Basterfield
135 <bob@cemetery.homeunix.org>.
136
137 2. Made restore to chdir() back into the initial directory when
138 dumping core while aborting a comparision operation. The
139 previous behaviour was to write the corefile at the root of
140 the directory being compared, which could very well be
141 read only and preventing the corefile generation. Thanks
142 to Kenneth Porter <shiva@sewingwitch.com> for the bug report.
143
144 3. Silenced the failure to call fgetflags() when comparing an
145 entry which has no ext2 attributes (as in lsattr()).
146
147 4. Fix a brown paper bug in restore -C which broke restore and
148 caused modifications on the filesystem being compared
149 (directories containing a file with the same name as the
150 directory get renamed to RSTTMP...). Thanks to Kenneth Porter
151 <shiva@sewingwitch.com> for finding the bug and helping me
152 reproduce it.
153
154 5. Made restore -C force the -N flag (no writing allowed on
155 the disk) in order to prevent more bugs like the above one.
156
157 Changes between versions 0.4b37 and 0.4b38 (released January 7, 2005)
158 =====================================================================
159
160 1. Fix a couple of troff syntax bugs in the man pages.
161 Thanks to Eric Raymond <esr@thyrsus.com> for the patch.
162
163 2. Made restore use either libncurses or libtermcap, depending
164 on which one is available at configure time.
165
166 3. Fixed restore negative size display bug when comparing a
167 dump containing files over 2GB. Thanks to Steve Bonds
168 <sbonds@users.sourceforge.net> for the bug report.
169
170 4. Do not save directory entries to non-dumped inodes
171 (excluded from dump). This will eliminate the 'missing
172 file' warnings when doing 'restore -C'.
173
174 5. Fix dump crash when backuping a huge (2TB) filesystem,
175 due to a bogus calculation on the inode map size.
176 Thanks to Kevin B. Haines <K.B.Haines@rl.ac.uk> for
177 submitting the bug and testing the fix.
178
179 6. Fix a problem in restore where the final \0 in the symbolic
180 link names could have been lost, generating corrupt filenames.
181 Thanks to Kyle Wilson <kyle.wilson@amd.com> for reporting the
182 bug.
183
184 7. Implemented a hash list for the directory names in restore.
185 The linear list used before caused problems in interactive
186 restores when dealing with directories having thousands of
187 entries. Thanks to Brian Ristuccia <bristuccia@starentnetworks.com>
188 for reporting the bug.
189
190 8. Improved restore -C, this time including the directory
191 attributes into the comparision.
192
193 9. Made restore understand tapes containing EA/ACLs (which will
194 be dumped by the next version of dump). In this version
195 extended attributes on the tape are ignored, for full EA/ACL
196 support wait for the next version or try the experimental EA
197 patch.
198
199 Changes between versions 0.4b36 and 0.4b37 (released July 7, 2004)
200 ==================================================================
201
202 1. Added the --enable-staticz configure option which enables
203 dump and restore to be linked with static versions of
204 libz and libbz2 (and dynamic versions of all other libraries).
205 This will make Debian users happy, because libz and libbz2
206 were the only needed libraries living in /usr, all the
207 others live in /lib. In case of system emergency, it is
208 better not to have to rely on an extra filesystem.
209 Thanks to Bdale Garbee <bdale@gag.com> for the suggestion.
210
211 2. Fix compilation on (at least the Linux Debian port to) AMD64.
212 (<ext2fs/ext2_types.h> defines some types (__s64 and __u64)
213 that are also defined by <linux/types.h> (<asm/types.h>) and
214 they conflict).
215
216 3. Make dump's reading of the dumpdates file a bit more robust,
217 preventing dump from crashing when the dumpdates file has
218 been modified by hand.
219
220 4. Fixed some offset calculations in dump code which could
221 lead to "bread lseek errors" on large filesystems. Thanks
222 to Bruce Lowekamp <lowekamp@users.sourceforge.net> for
223 reporting this bug and debugging the issue.
224
225 5. Made dump use the blkid library when searching for devices
226 by label or uuid instead of dump's own routines.
227
228 6. Corrected a bug in dump where a wrong LABEL=... line in
229 /etc/fstab could prevent dump from dumping unrelated
230 filesystems. Thanks to Bruce Lowekamp
231 <lowekamp@users.sourceforge.net> for reporting the bug.
232
233 Changes between versions 0.4b35 and 0.4b36 (released April 21, 2004)
234 ====================================================================
235
236 1. Fixed dump compilation with old gcc versions. Thanks to
237 Mike Castle <dalgoda@users.sourceforge.net> for the patch.
238
239 2. Fixed some warnings (howmany, roundup, powerof2 redefined)
240 when compiling against a recent glibc version.
241
242 3. Fixed a bug in restore preventing the read of a dump tape
243 written with Solaris 7 ufsdump. Thanks to Patrick Higgins
244 <phiggins@transzap.com> for reporting the bug and providing
245 the test case.
246
247 4. Changed dump to enable the creation of volumes of different
248 sizes in a single run (make -B accept a list of values).
249 Patch contributed by Florian Zumbiehl <florz@gmx.de>.
250
251 5. Use the glibc provided minor() and major() macros instead
252 of our own bitmask implementation. This should be safe for
253 when the major/minor namespace will migrate to 32 bits.
254 Thanks to Zhang Jun <zhangjun@nanjing-fnst.com> for reporting
255 the bug.
256
257 6. Made explicit in the dump man page that dump will not create
258 a remote file, it will only write to an already existing one.
259
260 7. Another try at making size estimates better again.
261
262 8. Put back the inconditional running of the end-of-tape
263 script which was changed in 0.434 to be run only when -M
264 or multiple -f were NOT specified. Some users rely on this
265 feature even when it is combined with -M/-f.
266
267 9. Fixed restore when restoring huge backups (where rstdir...
268 temporary files are over 2GB). Thanks to
269 Raphael Jaffey <rjaffey@artic.edu> for reporting this,
270 debugging the issue and making the patch.
271
272 10. Made restore understand FreeBSD UFS2 tapes. Thanks to
273 David <vrtin@users.sourceforge.net> for submitting the bug
274 and providing a test case.
275
276 11. Made dump work with an arbitrary integer level (no more 10
277 levels only limitation). Thanks to Matthew
278 <msvincen@midway.uchicago.edu> for the patch.
279
280 Changes between versions 0.4b34 and 0.4b35 (released December 21, 2003)
281 =======================================================================
282
283 1. Added a note in the dump man page saying that the default
284 blocksize can be 32 if -d is used with a high density argument.
285 Thanks to Antonios Christofides <A.Christofides@itia.ntua.gr>
286 for the patch.
287
288 2. Fixed configure to correctly understand CPPFLAGS, CFLAGS,
289 LDFLAGS environment variables. Thanks to Arcady Genkin
290 <antipode@users.sourceforge.net> for reporting the bug.
291
292 3. Made -e/-E options of dump accept an unlimited number of inodes
293 to be excluded, rather than a hardcoded maximum. Thanks to
294 Dietrich Rothe <d-rothe@users.sourceforge.net> for the patch.
295
296 4. Updated the autoconf system to 2.50. Forced the -D_BSD_SOURCE
297 and -D_USE_BSD_SIGNAL defines in configure in order to solve
298 64bit build problems because quad_t is redifined with a
299 different signature. Thanks to Mike Harris <mharris@redhat.com>
300 for reporting this bug.
301
302 5. Made restore build on Solaris, making possible to
303 restore Linux's "enhanced" tapes. Thanks to Uwe Gohlke
304 <uwe@ugsoft.de> for the patch.
305
306 6. Made an extension in the dump tape format capable of saving
307 MacOSX specific inode extensions. Uwe Gohlke <uwe@ugsoft.de>
308 wrote the extension and contributed the restore code back
309 into this codebase. The same extension mechanism will be
310 used in the future to save ACLs...
311
312 7. Made rmt work correctly with regard to QFA and local/remote
313 files and tapes. The remote access will however work only
314 when the dump provided rmt version is used. If you want to
315 use another rmt server, please do not use the QFA feature.
316 Thanks to Uwe Gohlke <uwe@ugsoft.de> for the patch.
317
318 Changes between versions 0.4b33 and 0.4b34 (released April 18, 2003)
319 ====================================================================
320
321 1. Fixed the INSTALL file to reflect the actual install paths.
322 Thanks to David Ronis <ronis@ronispc.chem.mcgill.ca> for
323 reporting the bug.
324
325 2. Fixed the configure script to only check for headers presence
326 instead of trying to compile them. This should fix issues
327 with old build environments. Thanks to Kari Lempiainen
328 <kari@funky.pp.fi> for reporting the bug.
329
330 3. Fixed restore to correctly ignore sockets when comparing
331 a dump (as socket cannot be properly restored anyway). Thanks
332 to Gunther Reiszig <gunther@mit.edu> for reporting the bug.
333
334 4. Fixed restore to correctly access the archive file (-A argument)
335 even when using a remote tape. Thanks to Fabrice Bellet
336 <fabrice@bellet.info> for reporting the bug.
337
338 5. Fixed (again) handling of long (largefile) seeks in rmt.
339 Thanks to Fabrice Bellet <fabrice@bellet.info> for reporting
340 the bug.
341
342 6. Fixed restore corner case when dealing with large block sizes
343 dump is able to write now (-b 1024). Thanks to Fabrice Bellet
344 <fabrice@bellet.info> for reporting the bug.
345
346 7. Fixed a bug preventing dump to access a filesystem having
347 a label exactly 16 bytes in length. Thanks to <minduni@ti-edu.ch>
348 for reporting the bug.
349
350 8. Made dump store dump dates using explicit timezones, fixing a
351 problem with incremental dumps if the timezone is modified
352 between the dumps. Thanks to Philippe Troin <phil@fifi.org> for
353 the bug report and the patch.
354
355 9. Fixed a bug encountered when dumping individual files (not full
356 filesystems or directories) and dangling symbolic links happen
357 to be in the list of files. For as far as dump is concerned,
358 dangling symbolic links are allowed, and are dumped as is.
359 Thanks to Jin-su Ahn <jsahn@ee.snu.ac.kr> for reporting the
360 bug and providing the fix.
361
362 10. Fixed open and creation modes and permissions for QFA and
363 table-of-contents files in dump and restore. Thanks to
364 Philippe Troin <phil@fifi.org> for the patch.
365
366 11. Fixed the archive file descriptor handling enabling it to be 0.
367 This can happen in some cases when shell redirections are used.
368 Thanks to Philippe Troin <phil@fifi.org> for the patch.
369
370 12. Delayed the opening of archive file until after suid had been
371 dropped (fixing a possible security issue if dump is suid).
372 Thanks to Philippe Troin <phil@fifi.org> for the patch.
373
374 13. Fixed the 'S' command handling in the rmt client part.
375 Thanks to Philippe Troin <phil@fifi.org> for the patch.
376
377 14. Modified the end-of-tape script handling to print out statistics
378 (and stop the timer) before launching the eot script. Also, the eot
379 script does not get run anymore when using -M (which makes sense) or
380 when multiple tapes are listed on the command line
381 (-f tape0,tape1,tapen) (which also makes sense).
382 Thanks to Philippe Troin <phil@fifi.org> for the patch.
383
384 15. Relicensed dump/restore under the 'revised' BSD license, as per
385 ftp://ftp.cs.berkeley.edu/ucb/4bsd/README.Impt.License.Change.
386
387 16. Added LZO compression to dump. This new compression method has
388 the advantage of being super fast, thus not killing tape streaming
389 on slow machines. Thanks to Helmut Jarausch
390 <jarausch@igpm.rwth-aachen.de> for the patch and to
391 Markus Oberhumer <markus@oberhumer.com> for giving special permission
392 to include his miniLZO project (GPL licensed) in dump/restore.
393
394 17. Some small buffer overruns fixes in rmt. Thanks to Antonomasia
395 <ant@notatla.demon.co.uk> for reporting the bugs.
396
397 18. Added a special rmt version which can do encryption when writing
398 to tape. Read examples/encrypted_rmt/README for details on
399 how to enable and configure it. Thanks to Ken Lalonde
400 <ken@globalremit.com> for the patch.
401
402 19. Made dump work with 2.5 kernel end of tape early warning semantics.
403 Thanks to Kai Makisara <Kai.Makisara@kolumbus.fi> for the patch.
404
405 20. Fixed a bug which caused dump -w|-W not to work anymore, because
406 the fs_freq and fs_passno fields in /etc/mtab are always set
407 to 0 0. Thanks to Trent Piepho <xyzzy@speakeasy.org> for
408 reporting the bug.
409
410 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
411 =======================================================================
412
413 1. Added a note in the restore man page clarifying the question
414 'set the permissions on the current directory ?' asked by
415 restore at the end of treatment in -i and -x modes.
416
417 2. Fixed the endianess issues when reading compressed tapes.
418 Thanks to Dark Force <daq4th@users.sourceforge.net> for
419 reporting this bug and providing test cases.
420
421 3. Fixed the "ACL won't be dumped" warning message (which showed
422 an extra, unrelated error message). Thanks to Dragan Krnic
423 <dkrnic@lycos.com> for reporting this bug.
424
425 4. Made dump look first into /etc/mtab, then into /etc/fstab
426 when searching for filesystem paths. Also fixed some problems
427 caused by binding mounts or multiple block device mounts.
428 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
429 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
430 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
431 and #160305, Sourceforge bugs #618699 and #679832).
432
433 5. Made dump's -I option accept the value '0' meaning all the
434 read errors will be ignored. This can be useful when running
435 dump from unattended sessions (like cron jobs). Thanks to
436 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
437
438 6. Fixed the output of dump to indicate 'blocks' instead of
439 'tape blocks' in the various messages (blocks are always
440 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
441 made it clearly print the current blocksize at the start of
442 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
443 the suggestions.
444
445 7. Made rmt understand long (largefiles) seeks.
446
447 8. Fixed build with very old versions of libext2fs, where
448 EXT2_FT_* constants were undefined.
449
450 9. Made dump accept the dumpdates path on the command line
451 (-D file option) instead of using only the hardcoded one.
452 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
453 suggestion.
454
455 10. Enabled rmt, LFS, readline, QFA options by default in
456 ./configure. Updated the configure process (new versions
457 of config.guess, config.sub etc).
458
459 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
460 =======================================================================
461
462 1. Changed dump to use fcntl(F_SETLK) style locking instead
463 of flock() when locking the dumpdates file. With the old
464 locking scheme, a local user having read rights on the
465 dumpdates file could be able to do a Denial of Service attack
466 on dump. In order to lock the dumpdates file with the new
467 scheme, the user would need to have write access on the file.
468 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
469 reporting the bug (originally a bugtraq post).
470
471 2. Fixed interactive 'ls' which caused spurious errors warnings
472 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
473 <jorgen_ostling@users.sourceforge.net> for reporting this
474 bug.
475
476 3. Fixed dump's estimate when dealing with sparse inodes.
477
478 4. Modified dump to allow setting a blocksize bigger than 32kB
479 (raised the limit to 1024kB), because newer hardware needs
480 this for proper operation. Thanks to Dirk Traenapp
481 <dtraenapp@users.sourceforge.net> for reporting this.
482
483 5. Fixed a bug causing Dump to stop and report an error if an
484 inode number in the exclude file was followed by some amount
485 of whitespace. Thanks to Jeffrey Sofferin
486 <sofferin@users.sourceforge.net> for reporting this bug.
487
488 6. Fixed a bug which caused restore, in some particular cases,
489 to ask some 'scary' questions and leave a bunch of RSTTMP
490 directories behind when restoring incremental tapes. Thanks
491 to Philippe Troin <phil@fifi.org> for reporting this bug and
492 providing the test cases.
493
494 7. Changed the wording when inodes are excluded from dump:
495 replaced 'Added inode 7 to exclude list' with
496 'Excluding inode 7 (resize inode) from dump', as suggested
497 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
498
499 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
500 ===================================================================
501
502 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
503 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
504 <eros@lamel.bo.cnr.it> for reporting the bug.
505
506 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
507 ===================================================================
508
509 1. Made dump print out the ext2 logical block number in case of a read
510 error, which can be given as an argument to debugfs' ncheck command,
511 in order to find out the affected inode. Added note about this
512 usage in dump's man page.
513
514 2. Fixed a problem in restore when reading tapes written on big
515 endian machines with very old versions of dump. The patch was
516 contributed by George Helffrich <george@geology.bristol.ac.uk>.
517
518 3. Fixed the tape length calculation when using large tapes
519 and compression. Thanks to Georg Lippold
520 <g_lippold@sourceforge.net> for reporting the bug.
521
522 4. Added a new set of examples in dump_on_cd_2 directory, based
523 on dump_on_cd examples but somewhat enhanced, supporting DVD
524 media, and localized in english and german. Thanks to
525 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
526
527 5. Made dump save 32 bit UID/GID. Internally, this was achieved
528 by switching from the old BSD inode format to the new BSD
529 inode format, which means that the tape format was changed.
530 However, since all restore versions out there should
531 transparently support both inode formats, the change should
532 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
533 reporting the bug.
534
535 6. Fixed a lot of warnings in the code shown when compiling
536 with 'gcc -W'. Thanks to Matthias Andree
537 <matthias.andree@stud.uni-dortmund.de> for reporting this.
538
539 7. Fixed a small markup bug in the dump man page. Thanks to
540 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
541 patch.
542
543 8. Rewrote entirely the man pages using the tmac.an macro
544 package (Linux man page format) instead of the original BSD
545 format. They should be now cleaner and easier to modify.
546
547 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
548 ==================================================================
549
550 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
551 non Linux clients were misinterpreted. The description of the problem
552 (incompatible numbering in Linux mtio opcodes) is documented at
553 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
554 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
555 providing an excellent, cross-platform replacement for rmt in his
556 star package.
557
558 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
559 which caused restore to fail when the inode maps were not entirely
560 contained on the first volume (when using really small volumes or
561 when dumping a huge number of inodes).
562
563 3. Fixed a problem in dump, where files in subdirectories of directories
564 that have the nodump flag set (or being excluded with -e/-E) were
565 sometimes incorrectly dumped (instead of being excluded). The original
566 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
567 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
568 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
569 help testing the patch.
570
571 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
572 which makes one able to pipe the output of dump, by the net, to
573 a remote CD-burner server.
574
575 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
576 and use GNU's symbolic syntax over rmt instead of numerical values
577 to assure multiple platform compatibility.
578
579 6. Documented the -d option in restore.
580
581 7. Added a -v (verbose) mode to dump. For now it just prints the number
582 of the inode being dumped, but this could evolve in future versions
583 to include interesting debugging output.
584
585 8. Added a -o flag to restore, which automatically restores the current
586 directory permissions in -i or -x mode, without asking the operator
587 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
588 and Peter Samuel <peters@e-smith.com>.
589
590 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
591 ====================================================================
592
593 1. Fixed a bug in the QFA feature which could have caused
594 a bad tape positionning on restore (causing extra delays in
595 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
596
597 2. Added a small note in the dump man page specifying that
598 there should be no white space between the option letter and
599 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
600
601 3. Made restore work with older versions of the readline library,
602 by checking in configure for several extended symbols. Restore
603 can now be compiled with a readline as old as the 2.0 release
604 (though it may be a good idea to upgrade it to a more recent
605 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
606 reporting the build failures.
607
608 4. Fixed a performance problem with the QFA file creation in
609 dump, which made unnecessary seeks on the tape slowing down
610 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
611 reporting this issue.
612
613 5. Removed the inclusion of some kernel headers in the dump
614 source, which prevented the compile in some kernel/glibc
615 headers/architecture combination. Thanks to Bdale Garbee
616 <bdale@gag.com> for reporting the bug.
617
618 6. Added the appropriate error message when dump fails to
619 open the output file for writing. Thanks to Amith Varghese
620 <amithv@yahoo.com> for reporting this bug.
621
622 7. Made restore able to understand large Solaris ufsdump tapes
623 (containing inodes bigger than 4194304). Sun have introduced
624 an "extension" to the dump tape format when dealing with
625 those inodes, which was uncorrectly handled by Linux restore.
626 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
627 providing a test case.
628
629 8. Added the -m parameter to dump which optimises the output for
630 inodes having been changed but not modified since the last dump
631 ('changed' and 'modified' have the meaning defined in stat(2)).
632 For those inodes, dump will save only the metadata, instead of
633 saving the entire inode contents. Inodes which are either
634 directories or have been modified since the last dump are saved
635 in a regular way. Uses of this flag must be consistent, meaning
636 that either every dump in an incremental dump set have the flag,
637 or no one has it.
638
639 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
640 =======================================================================
641
642 1. Fixed behaviour of dump when exceeding resource limits
643 (SIGXFSZ treatment).
644
645 2. Added the -L flag to restore to allow the user to specify a
646 maximal allowed number of miscompares when using restore
647 with the -C option to check the backup.
648
649 3. Detailed the manual entry for the -N option of restore.
650
651 4. Added the -a flag to restore to make able doing unattended
652 restores in -i or -x mode (automatically walks through the
653 multiple dump volumes).
654
655 5. Extended the QFA mode to work with local files and/or
656 remote tapes and files. This way, restore can know in advance
657 the tape number and the offset for the inodes to extract and
658 can minimize the extraction time by seeking directly to the
659 good tape/offset.
660
661 6. Added the -A <archive> option to both dump and restore,
662 which makes dump to archive a dump table-of-contents in
663 the specified file to be used by restore to determine
664 whether a file is in the dump file that is being restored.
665 (the archive file syntax is also compatible with the
666 Solaris ufsdump generated one).
667
668 7. Small fix in restore making it able to read some (broken ?)
669 Solaris ufsdump tapes.
670
671 8. Fixed dump to correctly recognise the root filesystem when using
672 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
673 for reporting this bug.
674
675 9. Added the -P <file> option to restore to create a
676 Quick File Access file from an already made dump. Patch
677 contributed by Uwe Gohlke <uwe@ugsoft.de>.
678
679 10. Made restore compile and run on Solaris, making it a
680 possible replacement for the standard ufsrestore. Port was
681 contributed by Uwe Gohlke <uwe@ugsoft.de>.
682
683 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
684 =====================================================================
685
686 1. Added a set of backup scripts from Eugenio Diaz
687 <getnito@yahoo.com> in the examples section. It features
688 automatic (cron based) full and incremental dumping of
689 several filesystems on a separate filesystem.
690
691 2. Fixed a off-by-one miscalculation which disabled dumping
692 a one letter subdirectory. Thanks to Chris Siebenmann
693 <cks@utcc.utoronto.ca> for reporting the bug.
694
695 3. Fixed several restore bugs occuring when trying to
696 restore some missing files on the tape. Thanks to Chris
697 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
698
699 4. Fixed --with-ldopts configure argument passing, installing from
700 a separate object directory, makefile cleanups contributed
701 by <splite@purdue.edu>.
702
703 5. Fix a bug which could caused, in some conditions, the highest
704 number inode of a filesystem, to not be dumped. Many thanks
705 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
706 chase this bug.
707
708 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
709 =======================================================================
710
711 1. Added a mini howto from Patrick Walsh in the examples
712 directory.
713
714 2. Minor man pages syntax corrections. Thanks to
715 Chris Danis <screechco@home.com> for reporting the bugs.
716
717 3. Added a script from David B. Peterson <dave@toppledwagon.com>
718 to the examples section. It features dumping several
719 filesystems to a remote tape drive upon ssh.
720
721 4. Added a patch provided by Richard Jones <rich@annexia.org>
722 which allows BRADEMAX (number of read errors tolerated by
723 dump) to be adjusted using the -I option.
724
725 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
726 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
727 patch.
728
729 6. Add the -l option to restore to specify if, when doing a
730 remote restore, the file used to access the backup is a
731 regular file (the defaults being a tape drive). Restore needs
732 to know this information when reading a remote compressed
733 dump. Previously, this information was autodetected, but
734 the autodetection code fails (with ioctl: Inappropriate ioctl
735 for device) when using a non Linux remote box. Thanks to
736 many users and especially to Eros Albertazzi
737 <eros@lamel.bo.cnr.it> for reporting this.
738
739 7. Found a workaroung for the dump deadlock problem (3 childs
740 stuck in pause(), father in read()). The workaround seems
741 to work for me and several beta-testers. If it doesn't work
742 for you, please report back.
743
744 8. Updated the RPM spec file (BuildPrereq, URL etc).
745
746 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
747 ========================================================================
748
749 1. Fixed the permissions of a newly created QFA file by dump.
750
751 2. Cleaned up the source of dump (the external variables
752 definition was a complete mess, making possible to have
753 objects overlap).
754
755 3. Fixed restore to use the full tape volume path when doing
756 a compare (since it changes the working directory to the
757 filesystem being compared in the process).
758
759 4. Added the -q option to dump which makes dump abort
760 whenever it needs operator attention. It should be
761 easier to use dump in scripts with this option.
762
763 5. Detect the use of incompatible options to dump and
764 refuse them (like -a and -B options together).
765
766 6. Added bzip2 compression to dump/restore (use option -j level
767 to select it). Note that you will need a restore version
768 >= 0.4b24 in order to restore a bzip2 compressed dump.
769 The same warning as for the zlib compression applies:
770 the tape format of a bzip2 dump is not compatible with the
771 original BSD tape format.
772
773 7. Fixed a overflow problem in dump corrupting the dump when
774 very large files were encountered. Thanks to Vladimir Nadvornik
775 for the bug report.
776
777 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
778 kernel buffer/page cache before starting the dump, helping
779 a bit those who use dump on mounted filesystems. Thanks to
780 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
781 for this suggestion.
782
783 9. Updated the RPM spec file following the RedHat changes
784 (dynamically linked binaries now in /usr/sbin etc).
785
786 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
787 which enables restore to recognise multi volume compressed dumps
788 done on CD. Included his scripts for dump (which pipe the dump
789 output directly into cdrecord) and restore. There is now
790 possible to to backups to CD on the fly!
791
792 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
793 ===================================================================
794
795 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
796 Marcus Meissner (Caldera International Security Dept.).
797
798 2. Implement the Sun rmt extended protocol. Patch submitted
799 by Ian Gordon <iangordon@users.sourceforge.net>.
800
801 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
802 the linux kernel header. This ensures that dump always has the
803 latest version of this file. Patch submitted by
804 Andreas Dilger <adilger@turbolinux.com>.
805
806 4. Report any filesystem present in either /etc/fstab with a
807 positive passno or /etc/dumpdates in dump -w output.
808 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
809
810 5. Fixed the looping problem in dump introduced in the
811 previous version.
812
813 6. Changed the -B option of dump to limit the size of
814 _compressed_ output per volume if compression is on.
815 Patch contributed by Helmut Jarausch
816 <jarausch@igpm.rwth-aachen.de>. Note however that, since
817 it is impossible to predict the size of the compressed
818 data before doing the compression, the -B limit is a bit
819 conservative.
820
821 7. Fixed a bug in reading the operator typed file/tape path for
822 the next volume in restore.
823
824 8. Implemented a "-F script" option to restore which permits the
825 user to specify a script which will be launched at the
826 beginning of each tape, useful for automatic programming of
827 tape changers for example. See the restore man page for the
828 script parameters and return codes.
829
830 9. Small fix for the QFA routines provided by Uwe Gohlke
831 <uwe@ugsoft.de>, and some recommendations for QFA uses in
832 the man pages.
833
834 10. Fixed the multivolume restoring where making a mistake
835 to the 'Mount next tape' prompt caused several blocks to
836 be lost.
837
838 11. Enhanced the -e option of dump to take as a parameter a
839 comma separated list of inode numbers.
840
841 12. Added the -E option to dump which specify a file containing
842 inode numbers to exclude from the dump.
843
844 13. Fixed the compressed multi-volume dump + restore.
845
846 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
847 ==================================================================
848
849 1. Made dump report the number of blocks dumped per volume.
850 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
851
852 2. Fix a bug in dump where waiting too much at the 'change volumes'
853 question caused the volume to be rewritten. Thanks to
854 Graham Reed <greed@users.sourceforge.net> for reporting the
855 bug and providing a patch.
856
857 3. Added a compression option to dump, contributed by Jerry
858 Peters <gapeters@worldnet.att.net>.
859
860 WARNING: the compression works only on files/pipes and
861 drives supporting variable block size.
862
863 WARNING2: tapes created using the compression feature are
864 incompatible with the standard dump tape format, and a
865 version of restore >= 0.4b22 will be needed for extraction.
866
867 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
868 architectures. Thanks to Paul Slootman <paul@debian.org> for
869 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
870 it upstream.
871
872 5. Many cleanups (CPP defines, const char warnings, check of
873 ext2fs COMPAT flags, time_t cleanups, added libext2 version
874 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
875
876 6. Made --prefix option work in configure. All the install path
877 are now based on the configure parameters.
878
879 7. Added the Quick File Access mode in dump/restore, contributed
880 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
881 a file tape position for each inode, and this file is used by
882 restore (if called with parameter Q and the filename)
883 to directly position the tape at the file restore is currently
884 working on. This saves hours when restoring single files from
885 large backups, saves the tapes and the drive's head. Use
886 --enable-qfa option of configure to compile in the QFA support.
887
888 8. Added the possibility to dump several files and directories
889 in a single invocation of dump. Thanks to Uwe Gohlke
890 <uwe@ugsoft.de> for implementing this option.
891
892 9. Fixed the dumping and restoring of files with holes
893 (files ending with holes didn't get dumped or restored
894 correctly).
895
896 10. Fixed a socket descriptor leak in dump, which leaved opened
897 3 file descriptors per dump process (and there is one dump
898 process per tape).
899
900 11. Fixed dump large file system support, by explicit use of
901 open64/lseek64/etc functions (explicit use needed because
902 e2fsprogs libraries don't behave well when compiled with
903 FILE_OFFSET_BITS=64).
904
905 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
906 ======================================================================
907
908 1. Fixed some bugs in the dump -w|-W logic introduced by
909 the previous version. Thanks to Andreas Dilger
910 <adilger@turbolinux.com> for his help on this one.
911
912 2. Fixed again a compilation problem when using old e2fs
913 headers (filesystem label related). Thanks to many users
914 who reported this stupid error.
915
916 3. Fixed a build problem on old lib5 based systems dealing with
917 _PATH_MNTTAB being undefined. Thanks to John Adams
918 <johna@onevista.com> for reporting the bug.
919
920 4. Improved the error detection in restore when creating
921 the temporary files in TMPDIR. Now restore will corectly
922 report a 'no space left on device' error instead of
923 strange errors which could imply an archive corruption.
924 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
925 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
926
927 5. Added the throughput information in dump's progression
928 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
929 for the patch.
930
931 6. Use libext2fs's inode scanning routines, which are particularly
932 robust when dealing with errors (especially when having some
933 bad blocks on the filesystem). Many thanks to Colin
934 <colin@horizon.com> for reporting the bug, investigating
935 the issues, finding a workaround, writing the patch and
936 fully testing it... (of course, if this patch will break
937 anything, he is to blame for :-)).
938
939 7. Made dump and restore LFS aware. Dump can dump a filesystem
940 containing large files, generate a large file on output and
941 restore can restore them. This feature is not enabled by
942 default, use --enable-largefile option of configure to enable
943 it (you need to have a LFS aware glibc though). Thanks to
944 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
945 and to Theodore T'so <tytso@valinux.com> for his always
946 useful thoughts.
947
948 8. Made dump ask upon a tape write error if it should rewrite
949 the current volume (assume this is a bad tape) or if it should
950 assume an end-of-tape condition (useful for tape drives which
951 behaves badly at the end of the tape). Thanks to Andreas
952 Dilger <adilger@turbolinux.com> for the suggestion.
953
954 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
955 =======================================================================
956
957 1. Fixed a small compilation problem due to a change
958 in the definintion of the struct sigaction in
959 glibc 2.0 / libc5. Thanks to Gunther Schlegel
960 <schlegel@riege.de> for reporting the bug and to
961 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
962 a fix.
963
964 2. Modified the label and uuid analysis in order to be
965 self-contained (independant of kernel/libc headers). This
966 should fix the compile with older kernel/libc headers and
967 will preserve the functionality. Thanks to Bernhard Erdmann
968 <bernhard.erdmann@gmx.de> for reporting the bug.
969
970 3. The 'exclude inode' option, if applied to a directory
971 inode, excludes now correctly all the leaves of this
972 directory. Thanks to John R. Dennison
973 <gerdesas@users.sourceforge.net> for reporting the bug.
974
975 4. Fixed the '-e' option to disable the possibility
976 to exclude the root inode (which causes the dumps to
977 be unreadable by restore). Prevented array overflow
978 when multiple -e options are used.
979
980 5. Fixed dump to correctly interpret a filesystem argument
981 which is a mountpoint and it is not an absolute path
982 (as specified in the fstab). Thanks to Bernhard R. Erdmann
983 <be@berdmann.de> for reporting the bug.
984
985 6. Made dump able to backup files larger than 2 GB. Note that
986 dump still doesn't cope with files larger than 4 GB.
987
988 7. Restore the real uid and gid before invoking an external
989 RSH program (big hole when dump or restore is suid root!).
990
991 8. Get the values for the default tape device and the location
992 of fstab file from the system headers. Thanks to
993 Andreas Dilger <adilger@turbolinux.com> for the patch.
994
995 9. Made dump -w|-W report all recognized filesystems
996 present in either /etc/fstab or /etc/dumpdates, and present
997 the list in the same order as in fstab file. Thanks
998 to Andreas Dilger <adilger@turbolinux.com> for the patch.
999
1000 10. Made dump's -a (automatic end of tape detection) the
1001 default. Specifying one of -B, -c, -d or -s options will
1002 override the EOM detection. Thanks to Andreas Dilger
1003 <adilger@turbolinux.com> for the patch.
1004
1005 11. Save the ext2 filesystem label into the dump volume label.
1006 Specifying a volume label on the command line (-L option)
1007 will override this feature. Thanks to Andreas Dilger
1008 <adilger@turbolinux.com> for the patch.
1009
1010 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
1011 =====================================================================
1012
1013 1. Fixed the signal handling in dump (which I broke in 0.4b17)
1014 which was causing several strange dump failures (dump
1015 hanged or segmentation faults etc).
1016
1017 2. Specified the default blocksize in the dump man page.
1018
1019 3. Changed two info messages of restore to be written on stdout
1020 instead of stderr in order to leave stderr only for errors
1021 or warnings. Thanks to Stephen Atwell
1022 <satwell@urbana.css.mot.com> for the suggestion.
1023
1024 4. Corrected an off by one calculation which prevented
1025 dumping a directory having a 1 character name length.
1026 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
1027 for reporting the bug.
1028
1029 5. Reinforce dump in order to allow only level 0 and no
1030 -u option when dumping a subdirectory, and document
1031 this in the man page. Thanks to Bernhard Erdmann
1032 <bernhard.erdmann@gmx.de> for reporting the bug.
1033
1034 6. Fixed a small harmless bug in restore which caused
1035 in some conditions hard links to be restored several
1036 times (therefore generation some warning messages).
1037 Thanks to Kenneth Porter <shiva@well.com> for
1038 reporting the bug.
1039
1040 7. Updated the RPM spec file to the latest RedHat version,
1041 providing FHS packaging and other cosmetic changes.
1042 You will need RPM version >= 3.0.5 in order to build the RPM.
1043
1044 8. Updated the configure script to check for libtermcap
1045 before checking for libreadline, because we need this
1046 library in order to compile the readline support.
1047
1048 9. Made dump understand the LABEL= and UUID= notation
1049 both in /etc/fstab and on the command line. Note that
1050 you will need the /proc filesystem in order to use
1051 these notations. Thanks to Erik Troan <ewt@redhat.com>
1052 for providing the patch.
1053
1054 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
1055 ===================================================================
1056
1057 1. Fixed a potential buffer overflow in restore. Thanks
1058 to Stan Bubrouski <satan@fastdial.net> for reporting
1059 the bug.
1060
1061 2. Fixed a readline-related bug which prevented
1062 'cat DUMPFILE | restore -i -f -' from working. Thanks
1063 to Charles Karney <karney@users.sourceforge.net>
1064 for the bug report.
1065
1066 3. Changed a few "panic" into "exit", causing restore to
1067 be more stable against some attacks (like the last one
1068 reported on Bugtraq, although the last version of restore
1069 was not vulnerable - just dumped core). Thanks to
1070 Andreas Hasenack <andreas@conectiva.com.br> for reporting
1071 the bugs to me.
1072
1073 4. Removed the suid-root bit on dump and restore in the
1074 default build (and generated RPMs). It should be safer
1075 now. Users who need the suid-root capabilities in order
1076 to make network backups should read first the man page
1077 of dump and enable the suid bit manually.
1078
1079 5. Added -ltermcap to the compile parameters for restore
1080 when using readline support, in order to make the compile
1081 process work on some systems (SuSE ?). Thanks to
1082 Patrik Schindler <poc@pocnet.net> for reporting the bug.
1083
1084 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
1085 ====================================================================
1086
1087 1. The -F script is called now *only* at the end of a tape,
1088 not at the end of the dump. Thanks to Christian Haul
1089 <haul@informatik.tu-darmstadt.de> for the bug report.
1090
1091 Normally, the device name and the current volume number
1092 are passed on the command line of the script. However,
1093 if you want the old style script (with no arguments
1094 passed on the command line), you can enable it in
1095 configure with the --enable-oldstylefscript.
1096
1097 2. Use posix signal handling to preserve dump functionality
1098 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
1099 the patch.
1100
1101 3. Made the exit code of restore in case of a 'C'ompare
1102 command reflect the result of the compare. An exit status
1103 of 0 means the dump archive is correct, 1 means tape errors,
1104 2 means that some files were modified. Thanks to Kenneth Porter
1105 <shiva@well.com> for the suggestion.
1106
1107 4. Made (finally) quotes work in interactive restore.
1108
1109 5. Small fixes in order to allow dump to compile with a
1110 really old e2fsprogs version. Thanks to Ian Zimmerman
1111 <itz@speakeasy.org> for the bug report.
1112
1113 6. Add GNU readline capabilities to interactive restore.
1114 Use configure's --enable-readline flag to enable this feature.
1115 Thanks to Patrik Schindler <poc@pocnet.net> for the
1116 suggestion.
1117
1118 7. Do the compare on the fly in restore 'C' mode (this will
1119 allow not to exhaust the available /tmp space when
1120 ccmparing large files). Thanks to Kenneth Porter
1121 <shiva@well.com> for the suggestion.
1122
1123 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
1124 ====================================================================
1125
1126 1. Fixed some several duplicate 'DUMP: DUMP:' in the
1127 output of dump.
1128
1129 2. Corrected the estimation of blocks to dump. Note that
1130 this is still wrong for large directory inodes, due
1131 to the size of a BSD directory entry on the tape
1132 (estimation is based on the size of the ext2 directory,
1133 which is smaller because it doesn't contain the null
1134 character at the end).
1135
1136 3. Made dump report the total number of MB written to
1137 tape at the end of run. Thanks to W. Reilly Cooley
1138 <wcooley@nakedape.cc> for the patch.
1139
1140 4. Added the -X option to restore giving the possibility
1141 to read the names of the files to be extracted/listed
1142 from a text file (in addition of the command line).
1143 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
1144 patch.
1145
1146 5. Added the device name and the current volume number
1147 as arguments to the end of tape script (dump -F option).
1148
1149 6. Made the multi-volume dump work again (after having
1150 broken it in 0.4b15).
1151
1152 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
1153 ===================================================================
1154
1155 1. Added a prompt command in interactive restore mode. Thanks
1156 to Andreas Dilger <adilger@home.com> for the patch.
1157
1158 2. Fixed a buffer overflow problem in dump (caused by
1159 not checking the size of the filesystem parameter).
1160 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
1161 reporting this on Bugtraq (and to several dump users
1162 who forwarded me his mail).
1163
1164 3. Added the '-F script' option to dump in order to
1165 launch a script at the end of each tape (to be used
1166 with a tape changer, or to notify the sysadmin by
1167 pager etc.).
1168
1169 4. Fixed a bug in restore compare code caused by the changes
1170 I made in 0.4b14.
1171
1172 5. Fixed the treatment of options using the old BSD syntax
1173 in both dump and restore.
1174
1175 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
1176 =======================================================================
1177
1178 1. Fixed a bug in dump which may have caused invalid deleted
1179 directories to be dumped out if they were deleted after the
1180 mapping pass. This could occure on active filesystem and lead
1181 to heap corruption (causing dump malfunction in many possible ways).
1182 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
1183 bug and submitting the patch.
1184
1185 2. Fixed the handling of the filesystem parameter in dump. A
1186 '/mnt/disk' parameter caused the disk contents to be dumped,
1187 but a '/mnt/disk/' parameter caused the mountpoint directory
1188 to be dumped (generally an empty directory).
1189
1190 3. Improved the output of dump in order to tell which directory
1191 it is currently dumping (when dumping a subtree).
1192
1193 4. Added the '-e' exclude inode option to dump. Thanks to
1194 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
1195
1196 5. Added a REPORTING-BUGS file in order to provide a guide
1197 on how to correctly report a bug in dump/restore.
1198
1199 6. Fixed a restore problem when restoring a hard link to an inode
1200 having the immutable or append-only attribute set. Thanks to
1201 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
1202
1203 7. Fixed a compatibility problem between dumps made on little
1204 endian machines (the format was correct) and big endian
1205 machines (the format was incorrect). This fix break the
1206 compatibility with the older dumps made on big endian
1207 machines (sparc, m86k, ppc etc). For the first time in
1208 linux dump's history (I believe), the dumps made by *BSD,
1209 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
1210 so interchangeable. Thanks to Rob Cermak
1211 <cermak@ahab.rutgers.edu> for submitting the bug and
1212 helping me test the fix.
1213
1214 8. Fixed the way dump reports the remaining percent/time, if
1215 the number of blocks actually dumped exceeds the estimated
1216 number of blocks. Thanks to Jean-Paul van der Jagt
1217 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
1218
1219 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
1220 ======================================================================
1221
1222 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
1223 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
1224
1225 2. Removed the requirement to build the RPM as root from the
1226 spec file. Thanks to Christian Weisgerber
1227 <naddy@mips.rhein-neckar.de> for submitting this.
1228
1229 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
1230 causing dump to block when dumping really huge filesystems.
1231 Many thanks to Patrik Schindler <poc@pocnet.net> for
1232 helping me find this bug.
1233
1234 4. Fixed the treatment for an interrupt signal when dump access
1235 the remote tape through RSH. Thanks to Christian Weisgerber
1236 <naddy@mips.rhein-neckar.de> for providing the patch.
1237
1238 5. Fixed a bug which was causing dump/restore to display
1239 garbage characters instead of the remote host name.
1240
1241 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
1242 =====================================================================
1243
1244 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
1245 <kukuk@suse.de> for submitting the patch.
1246
1247 2. Fix for the exit code when using the size estimate option of
1248 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
1249 submitting the patch.
1250
1251 3. Handle EINTR in atomical reads/writes in dump, which was causing
1252 dump to fail on some systems. Thanks to Eric Jergensen
1253 <eric@dvns.com> for reporting the bug and submitting the patch.
1254
1255 4. Handle more than 16 characters for the device names in dumpdates.
1256 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
1257 tracking down the problem and proposing the solution.
1258
1259 5. Fixed a bug in dump which prevented the creation of the
1260 dumpdates file when doing a 0-level dump without already
1261 having a dumpdates file. Thanks to Patrik Schindler
1262 <poc@pocnet.net> for reporting the bug.
1263
1264 6. Changed the way dump 'S' flag reports the size estimate
1265 from number of blocks into bytes (making it compatible
1266 with the Solaris version, and simplifying things for
1267 amanda users). Thanks to Jason L Tibbitts III
1268 <tibbs@math.uh.edu> for reporting the bug.
1269
1270 7. Fixed a compatibility problem in linux/alpha dump tape format.
1271 Now the linux/alpha dump are (again) compatible with the
1272 other dump formats. But this breaks compatibility with
1273 older dumps made on alpha. Thanks to Mike Tibor
1274 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
1275
1276 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
1277 ======================================================================
1278
1279 1. Added a '--enable-kerberos' to configure.
1280
1281 2. Added a 'S' option to dump which determines the amount of space
1282 that is needed to perform the dump without actually doing it, similar
1283 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
1284 <cermak@ahab.rutgers.edu>.
1285
1286 3. Added a 'M' multi-volume option to dump and restore which enables
1287 dumping to multiple files (useful when dumping to an ext2
1288 partition to make several dump files in order to bypass the 2GB
1289 file size limitation). The 'f' argument is treated as a prefix and
1290 the output files will be named <prefix>001, <prefix>002 etc. With
1291 the 'M' flag, restore automatically selects the right file without
1292 asking to enter a new tape each time.
1293
1294 4. Fixed a memory leak which was causing dump to grow very big
1295 (270MB when dumping a 10GB filesystem...). Thanks to Jason
1296 Fearon <jasonf@netrider.org.au> for reporting the bug.
1297
1298 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
1299 ======================================================================
1300
1301 1. Make configure test if the system glob routines support
1302 extended syntax (ALTDIRFUNC). If not, use the internal glob
1303 routines instead of system ones. Thanks to Bernhard Sadlowski
1304 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
1305 and helping me resolve this and other minor libc5 compiling
1306 glitches.
1307
1308 2. Fix a problem when dumping a ext2fs with the 'filetype'
1309 feature enabled. Thanks to Patrick J. LoPresti
1310 <patl@cag.lcs.mit.edu> for reporting the bug and to
1311 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
1312
1313 3. Made the nodump flag work on directories. A directory which
1314 has the nodump flag gets never dumped, regardless of its
1315 contents.
1316
1317 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
1318 which allows dump on an active ext3 filesystem. However, this
1319 is a "quick and dirty" patch which enables backup of an ext3
1320 filesystem through the ext2 compatibility (by ignoring the
1321 NEEDS_RECOVERY bit). The journal file is not recognized and
1322 it is dumped (it should not).
1323
1324 5. Test the superblock compatibility flags when dumping, in order
1325 to be sure that we know how to deal with specific features.
1326
1327 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
1328 ====================================================================
1329
1330 1. Use lchown instead of chown, fixing a possible security problem
1331 when restoring symlinks (a malicious user could use this
1332 to deliberately corrupt the ownership of important system files).
1333 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
1334 this and providing the patch.
1335
1336 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
1337 ====================================================================
1338
1339 1. Put dump sources under CVS, added Id tags in all files so
1340 one can use 'ident' on binary files.
1341
1342 2. Added the dump/restore version in the usage text so one can
1343 easily verify the version he is using.
1344
1345 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
1346 a va_start/va_end problem on linux-ppc (always call va_start
1347 va_end in pairs each time we use a vararg function).
1348
1349 4. Added again the DT_* constants because old libc does not
1350 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
1351 for submitting the bug report.
1352
1353 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
1354 this should enable dumping big (huge) filesystems.
1355
1356 6. Added the RSH environment variable in order to be able to
1357 use a rsh replacement like ssh when doing remote backups (and
1358 bypass the security limitations of rcmd). Now you can do remote
1359 backups without being root (or making dump setuid root).
1360
1361 7. Modified again the way dumpdates works. For incremental dumps,
1362 we need to read dumpdates even if we are not using 'u' option.
1363 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
1364 this should work.
1365
1366 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
1367 ===================================================================
1368
1369 1. Removed the 'k' flag from the restore 'about' text if kerberos
1370 was not compiled in.
1371
1372 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
1373 to them (fsetflags takes a char*, setflags an open fd!).
1374
1375 3. (f)setflags is called only if the flags aren't empty. If the
1376 file is a special file, a warning is printed, because changing
1377 flags implies opening the device. Normally, a special file
1378 should not have any flag... (Debian bug #29775, patch provided
1379 by Abhijit Dasgupta <abhijit@ans.net>).
1380
1381 4. Made possible to dump a file system not mentioned in /etc/fstab.
1382 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
1383
1384 5. Changed the default behaviour to not create dumpdates
1385 unless 'u' option is specified. Removed the old "debian-patch"
1386 which provided the same thing. (Debian bug #38136, #33818).
1387
1388 6. Removed all those dump*announce, since they were getting old...
1389
1390 7. Added warning messages if dumpdates does not exist and
1391 when an inferior level dump does not exist (except for a level 0
1392 dump).
1393
1394 8. Debugged the glob calls in interactive mode: restore used a
1395 dirent struct which was different from the /usr/include/dirent.h
1396 one (this used to work, is it a glibc2 change?), so none of the
1397 compat glob (which used /usr/include/dirent.h) or the system glob
1398 worked. Restore use now the system dirent (and the system
1399 DT_* constants), which are compatible with BSD ones.
1400
1401 9. Added a configure flag (--with-dumpdatespath) to specify
1402 the location of dumpdates file. By default, it is
1403 /etc/dumpdates.
1404
1405 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1406 included the current date/version in man pages.
1407
1408 11. Corrected the estimation of remaining time when
1409 the operator doesn't change the tapes quickly enough. This
1410 was an old bug, I thought I corrected it, and discovered
1411 that in fact it was corrected in two different places, so
1412 the results canceled each other...
1413
1414 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1415 ===================================================================
1416
1417 1. Integrated multiple patches from RedHat, Debian and SuSE:
1418
1419 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1420 isn't specified.
1421 - several fixes in the man pages.
1422 - update the default tape device to /dev/st0.
1423 - many updates for Linux Alpha (byte ordering, size_t etc).
1424 - buffer overruns.
1425 - use environment variable for TMPDIR (instead of /tmp).
1426 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1427 - workaround egcs bug (RedHat bugs #4281 and #2989).
1428 - wire $(OPT) throughout Makefile's.
1429
1430 2. Upgrade the dump revision to 1, making possible to dump filesystems
1431 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1432
1433 3. Fix some compile warnings, prototype all functions.
1434
1435 4. Use glibc err/glob instead of internal compatibility
1436 routines (only if available).
1437
1438 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1439 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1440
1441 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1442 ======================================================================
1443
1444 1. Integrated the changes from FreeBSD-3.1-RELEASE
1445 (mostly bug fixes, buffer overruns, dump has now an "automatic
1446 tape length calculation" flag, dump/restore can use kerberos now
1447 (this is NOT tested), use environment variables for TAPE and
1448 RMT etc.).
1449
1450 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1451
1452 3. Corrected a bug in restore when using 'C' option with multi-volumes
1453 tapes (files splited accros two tapes give "size changed" errors
1454 when comparing).
1455
1456 4. Corrected the long standing bug when dumping multiple tapes.
1457 This works for me, needs further testing.
1458
1459 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1460 ====================================================================
1461
1462 1. Dump now runs correctly on kernels 2.1.x
1463 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1464
1465 Changes between versions 0.4b2 and 0.4b3
1466 ========================================
1467
1468 1. Use realpath() if available
1469
1470 2. Report statistics
1471
1472 Changes between versions 0.4b1 and 0.4b2
1473 ========================================
1474
1475 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1476 the patch)
1477
1478 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1479
1480 3. Fixed configure.in to correctly handle --enable-rmt
1481
1482 Changes between versions 0.3 and 0.4b1
1483 ======================================
1484
1485 1. Integrated the changes from 4.4BSD-Lite2
1486
1487 2. Integrated the patches from Debian and Red Hat
1488
1489 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1490
1491 4. Changed dump to use the Ext2fs library to get block addresses. This
1492 should solve the endianness problem on SparcLinux.
1493
1494 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1495 distribution's one) to use autoconf
1496
1497 6. Fixed a few minor bugs
1498
1499 Changes between versions 0.2e and 0.2f
1500 ======================================
1501
1502 1. Added the creation of named pipes (fifos) in restore.
1503
1504 2. Added the -N flag in the restore manual page.
1505
1506 3. Added the file kernel.patch which contains the llseek() optimization
1507 patch for 1.2.x kernels.
1508
1509 4. Fixed a bug in the restoration of symbolic links: owner and group were
1510 not restored.
1511
1512 5. Integrated some changes from FreeBSD 2.2.
1513
1514 6. Added a call to ftruncate() after restoring each file to restore
1515 correctly files ending by a hole.
1516
1517 Changes between versions 0.2d and 0.2e
1518 ======================================
1519
1520 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1521 with owner = root (0) and group = root/wheel/whatever (0).
1522
1523 Changes between versions 0.2c and 0.2d
1524 ======================================
1525
1526 1. Dump is now able to backup 2GB+ filesystems.
1527
1528 2. Dump and restore can now be linked as static binaries.
1529
1530 Changes between versions 0.2b and 0.2c
1531 ======================================
1532
1533 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1534
1535 Changes between versions 0.2a and 0.2b
1536 ======================================
1537
1538 1. Really fixed the bug that I should have corrected in 0.2a.
1539
1540 2. Enabled optimization again.
1541
1542 Changes between versions 0.2 and 0.2a
1543 =====================================
1544
1545 1. Disabled the optimization during compilation.
1546
1547 Changes between versions 0.1 and 0.2
1548 ====================================
1549
1550 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1551 the fs_type field (actually, I modified the file fstab.c to make it
1552 use the mntent functions).
1553
1554 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1555 made by dump should be readable by the BSD restore and backups made
1556 by the BSD dump should be readable by restore. Unfortunately, this
1557 means that the dump format has changed between version 0.1 and version
1558 0.2 :-(
1559
1560 3. Dump is now able to backup a subtree, it is no longer limited to whole
1561 filesystems like the BSD version.
1562
1563 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1564 than 2 GB.
1565
1566 Changes between versions 0.0 and 0.1
1567 ====================================
1568
1569 1. Now create links rdump and rrestore during the `make install' step.
1570
1571 2. Linux port specific bugs added to the manual pages
1572
1573 3. Incorrect estimation of the number of tapes blocks fixed when doing
1574 incremental backups.
1575
1576 4. Better ls-like format in restore in interactive mode.