[dump.git] / CHANGES
1 $Id: CHANGES,v 1.258 2004/11/22 10:22:40 stelian Exp $
3 Changes between versions 0.4b37 and 0.4b38 (released ????????????)
4 ==================================================================
6 1. Fix a couple of troff syntax bugs in the man pages.
7 Thanks to Eric Raymond <esr@thyrsus.com> for the patch.
9 2. Made restore use either libncurses or libtermcap, depending
10 on which one is available at configure time.
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.
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'.
20 Changes between versions 0.4b36 and 0.4b37 (released July 7, 2004)
21 ==================================================================
23 1. Added the --enable-staticz configure option which enables
24 dump and restore to be linked with static versions of
25 libz and libbz2 (and dynamic versions of all other libraries).
26 This will make Debian users happy, because libz and libbz2
27 were the only needed libraries living in /usr, all the
28 others live in /lib. In case of system emergency, it is
29 better not to have to rely on an extra filesystem.
30 Thanks to Bdale Garbee <bdale@gag.com> for the suggestion.
32 2. Fix compilation on (at least the Linux Debian port to) AMD64.
33 (<ext2fs/ext2_types.h> defines some types (__s64 and __u64)
34 that are also defined by <linux/types.h> (<asm/types.h>) and
35 they conflict).
37 3. Make dump's reading of the dumpdates file a bit more robust,
38 preventing dump from crashing when the dumpdates file has
39 been modified by hand.
41 4. Fixed some offset calculations in dump code which could
42 lead to "bread lseek errors" on large filesystems. Thanks
43 to Bruce Lowekamp <lowekamp@users.sourceforge.net> for
44 reporting this bug and debugging the issue.
46 5. Made dump use the blkid library when searching for devices
47 by label or uuid instead of dump's own routines.
49 6. Corrected a bug in dump where a wrong LABEL=... line in
50 /etc/fstab could prevent dump from dumping unrelated
51 filesystems. Thanks to Bruce Lowekamp
52 <lowekamp@users.sourceforge.net> for reporting the bug.
54 Changes between versions 0.4b35 and 0.4b36 (released April 21, 2004)
55 ====================================================================
57 1. Fixed dump compilation with old gcc versions. Thanks to
58 Mike Castle <dalgoda@users.sourceforge.net> for the patch.
60 2. Fixed some warnings (howmany, roundup, powerof2 redefined)
61 when compiling against a recent glibc version.
63 3. Fixed a bug in restore preventing the read of a dump tape
64 written with Solaris 7 ufsdump. Thanks to Patrick Higgins
65 <phiggins@transzap.com> for reporting the bug and providing
66 the test case.
68 4. Changed dump to enable the creation of volumes of different
69 sizes in a single run (make -B accept a list of values).
70 Patch contributed by Florian Zumbiehl <florz@gmx.de>.
72 5. Use the glibc provided minor() and major() macros instead
73 of our own bitmask implementation. This should be safe for
74 when the major/minor namespace will migrate to 32 bits.
75 Thanks to Zhang Jun <zhangjun@nanjing-fnst.com> for reporting
76 the bug.
78 6. Made explicit in the dump man page that dump will not create
79 a remote file, it will only write to an already existing one.
81 7. Another try at making size estimates better again.
83 8. Put back the inconditional running of the end-of-tape
84 script which was changed in 0.434 to be run only when -M
85 or multiple -f were NOT specified. Some users rely on this
86 feature even when it is combined with -M/-f.
88 9. Fixed restore when restoring huge backups (where rstdir...
89 temporary files are over 2GB). Thanks to
90 Raphael Jaffey <rjaffey@artic.edu> for reporting this,
91 debugging the issue and making the patch.
93 10. Made restore understand FreeBSD UFS2 tapes. Thanks to
94 David <vrtin@users.sourceforge.net> for submitting the bug
95 and providing a test case.
97 11. Made dump work with an arbitrary integer level (no more 10
98 levels only limitation). Thanks to Matthew
99 <msvincen@midway.uchicago.edu> for the patch.
101 Changes between versions 0.4b34 and 0.4b35 (released December 21, 2003)
102 =======================================================================
104 1. Added a note in the dump man page saying that the default
105 blocksize can be 32 if -d is used with a high density argument.
106 Thanks to Antonios Christofides <A.Christofides@itia.ntua.gr>
107 for the patch.
109 2. Fixed configure to correctly understand CPPFLAGS, CFLAGS,
110 LDFLAGS environment variables. Thanks to Arcady Genkin
111 <antipode@users.sourceforge.net> for reporting the bug.
113 3. Made -e/-E options of dump accept an unlimited number of inodes
114 to be excluded, rather than a hardcoded maximum. Thanks to
115 Dietrich Rothe <d-rothe@users.sourceforge.net> for the patch.
117 4. Updated the autoconf system to 2.50. Forced the -D_BSD_SOURCE
118 and -D_USE_BSD_SIGNAL defines in configure in order to solve
119 64bit build problems because quad_t is redifined with a
120 different signature. Thanks to Mike Harris <mharris@redhat.com>
121 for reporting this bug.
123 5. Made restore build on Solaris, making possible to
124 restore Linux's "enhanced" tapes. Thanks to Uwe Gohlke
125 <uwe@ugsoft.de> for the patch.
127 6. Made an extension in the dump tape format capable of saving
128 MacOSX specific inode extensions. Uwe Gohlke <uwe@ugsoft.de>
129 wrote the extension and contributed the restore code back
130 into this codebase. The same extension mechanism will be
131 used in the future to save ACLs...
133 7. Made rmt work correctly with regard to QFA and local/remote
134 files and tapes. The remote access will however work only
135 when the dump provided rmt version is used. If you want to
136 use another rmt server, please do not use the QFA feature.
137 Thanks to Uwe Gohlke <uwe@ugsoft.de> for the patch.
139 Changes between versions 0.4b33 and 0.4b34 (released April 18, 2003)
140 ====================================================================
142 1. Fixed the INSTALL file to reflect the actual install paths.
143 Thanks to David Ronis <ronis@ronispc.chem.mcgill.ca> for
144 reporting the bug.
146 2. Fixed the configure script to only check for headers presence
147 instead of trying to compile them. This should fix issues
148 with old build environments. Thanks to Kari Lempiainen
149 <kari@funky.pp.fi> for reporting the bug.
151 3. Fixed restore to correctly ignore sockets when comparing
152 a dump (as socket cannot be properly restored anyway). Thanks
153 to Gunther Reiszig <gunther@mit.edu> for reporting the bug.
155 4. Fixed restore to correctly access the archive file (-A argument)
156 even when using a remote tape. Thanks to Fabrice Bellet
157 <fabrice@bellet.info> for reporting the bug.
159 5. Fixed (again) handling of long (largefile) seeks in rmt.
160 Thanks to Fabrice Bellet <fabrice@bellet.info> for reporting
161 the bug.
163 6. Fixed restore corner case when dealing with large block sizes
164 dump is able to write now (-b 1024). Thanks to Fabrice Bellet
165 <fabrice@bellet.info> for reporting the bug.
167 7. Fixed a bug preventing dump to access a filesystem having
168 a label exactly 16 bytes in length. Thanks to <minduni@ti-edu.ch>
169 for reporting the bug.
171 8. Made dump store dump dates using explicit timezones, fixing a
172 problem with incremental dumps if the timezone is modified
173 between the dumps. Thanks to Philippe Troin <phil@fifi.org> for
174 the bug report and the patch.
176 9. Fixed a bug encountered when dumping individual files (not full
177 filesystems or directories) and dangling symbolic links happen
178 to be in the list of files. For as far as dump is concerned,
179 dangling symbolic links are allowed, and are dumped as is.
180 Thanks to Jin-su Ahn <jsahn@ee.snu.ac.kr> for reporting the
181 bug and providing the fix.
183 10. Fixed open and creation modes and permissions for QFA and
184 table-of-contents files in dump and restore. Thanks to
185 Philippe Troin <phil@fifi.org> for the patch.
187 11. Fixed the archive file descriptor handling enabling it to be 0.
188 This can happen in some cases when shell redirections are used.
189 Thanks to Philippe Troin <phil@fifi.org> for the patch.
191 12. Delayed the opening of archive file until after suid had been
192 dropped (fixing a possible security issue if dump is suid).
193 Thanks to Philippe Troin <phil@fifi.org> for the patch.
195 13. Fixed the 'S' command handling in the rmt client part.
196 Thanks to Philippe Troin <phil@fifi.org> for the patch.
198 14. Modified the end-of-tape script handling to print out statistics
199 (and stop the timer) before launching the eot script. Also, the eot
200 script does not get run anymore when using -M (which makes sense) or
201 when multiple tapes are listed on the command line
202 (-f tape0,tape1,tapen) (which also makes sense).
203 Thanks to Philippe Troin <phil@fifi.org> for the patch.
205 15. Relicensed dump/restore under the 'revised' BSD license, as per
206 ftp://ftp.cs.berkeley.edu/ucb/4bsd/README.Impt.License.Change.
208 16. Added LZO compression to dump. This new compression method has
209 the advantage of being super fast, thus not killing tape streaming
210 on slow machines. Thanks to Helmut Jarausch
211 <jarausch@igpm.rwth-aachen.de> for the patch and to
212 Markus Oberhumer <markus@oberhumer.com> for giving special permission
213 to include his miniLZO project (GPL licensed) in dump/restore.
215 17. Some small buffer overruns fixes in rmt. Thanks to Antonomasia
216 <ant@notatla.demon.co.uk> for reporting the bugs.
218 18. Added a special rmt version which can do encryption when writing
219 to tape. Read examples/encrypted_rmt/README for details on
220 how to enable and configure it. Thanks to Ken Lalonde
221 <ken@globalremit.com> for the patch.
223 19. Made dump work with 2.5 kernel end of tape early warning semantics.
224 Thanks to Kai Makisara <Kai.Makisara@kolumbus.fi> for the patch.
226 20. Fixed a bug which caused dump -w|-W not to work anymore, because
227 the fs_freq and fs_passno fields in /etc/mtab are always set
228 to 0 0. Thanks to Trent Piepho <xyzzy@speakeasy.org> for
229 reporting the bug.
231 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
232 =======================================================================
234 1. Added a note in the restore man page clarifying the question
235 'set the permissions on the current directory ?' asked by
236 restore at the end of treatment in -i and -x modes.
238 2. Fixed the endianess issues when reading compressed tapes.
239 Thanks to Dark Force <daq4th@users.sourceforge.net> for
240 reporting this bug and providing test cases.
242 3. Fixed the "ACL won't be dumped" warning message (which showed
243 an extra, unrelated error message). Thanks to Dragan Krnic
244 <dkrnic@lycos.com> for reporting this bug.
246 4. Made dump look first into /etc/mtab, then into /etc/fstab
247 when searching for filesystem paths. Also fixed some problems
248 caused by binding mounts or multiple block device mounts.
249 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
250 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
251 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
252 and #160305, Sourceforge bugs #618699 and #679832).
254 5. Made dump's -I option accept the value '0' meaning all the
255 read errors will be ignored. This can be useful when running
256 dump from unattended sessions (like cron jobs). Thanks to
257 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
259 6. Fixed the output of dump to indicate 'blocks' instead of
260 'tape blocks' in the various messages (blocks are always
261 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
262 made it clearly print the current blocksize at the start of
263 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
264 the suggestions.
266 7. Made rmt understand long (largefiles) seeks.
268 8. Fixed build with very old versions of libext2fs, where
269 EXT2_FT_* constants were undefined.
271 9. Made dump accept the dumpdates path on the command line
272 (-D file option) instead of using only the hardcoded one.
273 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
274 suggestion.
276 10. Enabled rmt, LFS, readline, QFA options by default in
277 ./configure. Updated the configure process (new versions
278 of config.guess, config.sub etc).
280 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
281 =======================================================================
283 1. Changed dump to use fcntl(F_SETLK) style locking instead
284 of flock() when locking the dumpdates file. With the old
285 locking scheme, a local user having read rights on the
286 dumpdates file could be able to do a Denial of Service attack
287 on dump. In order to lock the dumpdates file with the new
288 scheme, the user would need to have write access on the file.
289 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
290 reporting the bug (originally a bugtraq post).
292 2. Fixed interactive 'ls' which caused spurious errors warnings
293 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
294 <jorgen_ostling@users.sourceforge.net> for reporting this
295 bug.
297 3. Fixed dump's estimate when dealing with sparse inodes.
299 4. Modified dump to allow setting a blocksize bigger than 32kB
300 (raised the limit to 1024kB), because newer hardware needs
301 this for proper operation. Thanks to Dirk Traenapp
302 <dtraenapp@users.sourceforge.net> for reporting this.
304 5. Fixed a bug causing Dump to stop and report an error if an
305 inode number in the exclude file was followed by some amount
306 of whitespace. Thanks to Jeffrey Sofferin
307 <sofferin@users.sourceforge.net> for reporting this bug.
309 6. Fixed a bug which caused restore, in some particular cases,
310 to ask some 'scary' questions and leave a bunch of RSTTMP
311 directories behind when restoring incremental tapes. Thanks
312 to Philippe Troin <phil@fifi.org> for reporting this bug and
313 providing the test cases.
315 7. Changed the wording when inodes are excluded from dump:
316 replaced 'Added inode 7 to exclude list' with
317 'Excluding inode 7 (resize inode) from dump', as suggested
318 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
320 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
321 ===================================================================
323 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
324 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
325 <eros@lamel.bo.cnr.it> for reporting the bug.
327 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
328 ===================================================================
330 1. Made dump print out the ext2 logical block number in case of a read
331 error, which can be given as an argument to debugfs' ncheck command,
332 in order to find out the affected inode. Added note about this
333 usage in dump's man page.
335 2. Fixed a problem in restore when reading tapes written on big
336 endian machines with very old versions of dump. The patch was
337 contributed by George Helffrich <george@geology.bristol.ac.uk>.
339 3. Fixed the tape length calculation when using large tapes
340 and compression. Thanks to Georg Lippold
341 <g_lippold@sourceforge.net> for reporting the bug.
343 4. Added a new set of examples in dump_on_cd_2 directory, based
344 on dump_on_cd examples but somewhat enhanced, supporting DVD
345 media, and localized in english and german. Thanks to
346 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
348 5. Made dump save 32 bit UID/GID. Internally, this was achieved
349 by switching from the old BSD inode format to the new BSD
350 inode format, which means that the tape format was changed.
351 However, since all restore versions out there should
352 transparently support both inode formats, the change should
353 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
354 reporting the bug.
356 6. Fixed a lot of warnings in the code shown when compiling
357 with 'gcc -W'. Thanks to Matthias Andree
358 <matthias.andree@stud.uni-dortmund.de> for reporting this.
360 7. Fixed a small markup bug in the dump man page. Thanks to
361 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
362 patch.
364 8. Rewrote entirely the man pages using the tmac.an macro
365 package (Linux man page format) instead of the original BSD
366 format. They should be now cleaner and easier to modify.
368 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
369 ==================================================================
371 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
372 non Linux clients were misinterpreted. The description of the problem
373 (incompatible numbering in Linux mtio opcodes) is documented at
374 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
375 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
376 providing an excellent, cross-platform replacement for rmt in his
377 star package.
379 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
380 which caused restore to fail when the inode maps were not entirely
381 contained on the first volume (when using really small volumes or
382 when dumping a huge number of inodes).
384 3. Fixed a problem in dump, where files in subdirectories of directories
385 that have the nodump flag set (or being excluded with -e/-E) were
386 sometimes incorrectly dumped (instead of being excluded). The original
387 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
388 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
389 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
390 help testing the patch.
392 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
393 which makes one able to pipe the output of dump, by the net, to
394 a remote CD-burner server.
396 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
397 and use GNU's symbolic syntax over rmt instead of numerical values
398 to assure multiple platform compatibility.
400 6. Documented the -d option in restore.
402 7. Added a -v (verbose) mode to dump. For now it just prints the number
403 of the inode being dumped, but this could evolve in future versions
404 to include interesting debugging output.
406 8. Added a -o flag to restore, which automatically restores the current
407 directory permissions in -i or -x mode, without asking the operator
408 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
409 and Peter Samuel <peters@e-smith.com>.
411 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
412 ====================================================================
414 1. Fixed a bug in the QFA feature which could have caused
415 a bad tape positionning on restore (causing extra delays in
416 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
418 2. Added a small note in the dump man page specifying that
419 there should be no white space between the option letter and
420 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
422 3. Made restore work with older versions of the readline library,
423 by checking in configure for several extended symbols. Restore
424 can now be compiled with a readline as old as the 2.0 release
425 (though it may be a good idea to upgrade it to a more recent
426 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
427 reporting the build failures.
429 4. Fixed a performance problem with the QFA file creation in
430 dump, which made unnecessary seeks on the tape slowing down
431 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
432 reporting this issue.
434 5. Removed the inclusion of some kernel headers in the dump
435 source, which prevented the compile in some kernel/glibc
436 headers/architecture combination. Thanks to Bdale Garbee
437 <bdale@gag.com> for reporting the bug.
439 6. Added the appropriate error message when dump fails to
440 open the output file for writing. Thanks to Amith Varghese
441 <amithv@yahoo.com> for reporting this bug.
443 7. Made restore able to understand large Solaris ufsdump tapes
444 (containing inodes bigger than 4194304). Sun have introduced
445 an "extension" to the dump tape format when dealing with
446 those inodes, which was uncorrectly handled by Linux restore.
447 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
448 providing a test case.
450 8. Added the -m parameter to dump which optimises the output for
451 inodes having been changed but not modified since the last dump
452 ('changed' and 'modified' have the meaning defined in stat(2)).
453 For those inodes, dump will save only the metadata, instead of
454 saving the entire inode contents. Inodes which are either
455 directories or have been modified since the last dump are saved
456 in a regular way. Uses of this flag must be consistent, meaning
457 that either every dump in an incremental dump set have the flag,
458 or no one has it.
460 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
461 =======================================================================
463 1. Fixed behaviour of dump when exceeding resource limits
464 (SIGXFSZ treatment).
466 2. Added the -L flag to restore to allow the user to specify a
467 maximal allowed number of miscompares when using restore
468 with the -C option to check the backup.
470 3. Detailed the manual entry for the -N option of restore.
472 4. Added the -a flag to restore to make able doing unattended
473 restores in -i or -x mode (automatically walks through the
474 multiple dump volumes).
476 5. Extended the QFA mode to work with local files and/or
477 remote tapes and files. This way, restore can know in advance
478 the tape number and the offset for the inodes to extract and
479 can minimize the extraction time by seeking directly to the
480 good tape/offset.
482 6. Added the -A <archive> option to both dump and restore,
483 which makes dump to archive a dump table-of-contents in
484 the specified file to be used by restore to determine
485 whether a file is in the dump file that is being restored.
486 (the archive file syntax is also compatible with the
487 Solaris ufsdump generated one).
489 7. Small fix in restore making it able to read some (broken ?)
490 Solaris ufsdump tapes.
492 8. Fixed dump to correctly recognise the root filesystem when using
493 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
494 for reporting this bug.
496 9. Added the -P <file> option to restore to create a
497 Quick File Access file from an already made dump. Patch
498 contributed by Uwe Gohlke <uwe@ugsoft.de>.
500 10. Made restore compile and run on Solaris, making it a
501 possible replacement for the standard ufsrestore. Port was
502 contributed by Uwe Gohlke <uwe@ugsoft.de>.
504 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
505 =====================================================================
507 1. Added a set of backup scripts from Eugenio Diaz
508 <getnito@yahoo.com> in the examples section. It features
509 automatic (cron based) full and incremental dumping of
510 several filesystems on a separate filesystem.
512 2. Fixed a off-by-one miscalculation which disabled dumping
513 a one letter subdirectory. Thanks to Chris Siebenmann
514 <cks@utcc.utoronto.ca> for reporting the bug.
516 3. Fixed several restore bugs occuring when trying to
517 restore some missing files on the tape. Thanks to Chris
518 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
520 4. Fixed --with-ldopts configure argument passing, installing from
521 a separate object directory, makefile cleanups contributed
522 by <splite@purdue.edu>.
524 5. Fix a bug which could caused, in some conditions, the highest
525 number inode of a filesystem, to not be dumped. Many thanks
526 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
527 chase this bug.
529 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
530 =======================================================================
532 1. Added a mini howto from Patrick Walsh in the examples
533 directory.
535 2. Minor man pages syntax corrections. Thanks to
536 Chris Danis <screechco@home.com> for reporting the bugs.
538 3. Added a script from David B. Peterson <dave@toppledwagon.com>
539 to the examples section. It features dumping several
540 filesystems to a remote tape drive upon ssh.
542 4. Added a patch provided by Richard Jones <rich@annexia.org>
543 which allows BRADEMAX (number of read errors tolerated by
544 dump) to be adjusted using the -I option.
546 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
547 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
548 patch.
550 6. Add the -l option to restore to specify if, when doing a
551 remote restore, the file used to access the backup is a
552 regular file (the defaults being a tape drive). Restore needs
553 to know this information when reading a remote compressed
554 dump. Previously, this information was autodetected, but
555 the autodetection code fails (with ioctl: Inappropriate ioctl
556 for device) when using a non Linux remote box. Thanks to
557 many users and especially to Eros Albertazzi
558 <eros@lamel.bo.cnr.it> for reporting this.
560 7. Found a workaroung for the dump deadlock problem (3 childs
561 stuck in pause(), father in read()). The workaround seems
562 to work for me and several beta-testers. If it doesn't work
563 for you, please report back.
565 8. Updated the RPM spec file (BuildPrereq, URL etc).
567 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
568 ========================================================================
570 1. Fixed the permissions of a newly created QFA file by dump.
572 2. Cleaned up the source of dump (the external variables
573 definition was a complete mess, making possible to have
574 objects overlap).
576 3. Fixed restore to use the full tape volume path when doing
577 a compare (since it changes the working directory to the
578 filesystem being compared in the process).
580 4. Added the -q option to dump which makes dump abort
581 whenever it needs operator attention. It should be
582 easier to use dump in scripts with this option.
584 5. Detect the use of incompatible options to dump and
585 refuse them (like -a and -B options together).
587 6. Added bzip2 compression to dump/restore (use option -j level
588 to select it). Note that you will need a restore version
589 >= 0.4b24 in order to restore a bzip2 compressed dump.
590 The same warning as for the zlib compression applies:
591 the tape format of a bzip2 dump is not compatible with the
592 original BSD tape format.
594 7. Fixed a overflow problem in dump corrupting the dump when
595 very large files were encountered. Thanks to Vladimir Nadvornik
596 for the bug report.
598 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
599 kernel buffer/page cache before starting the dump, helping
600 a bit those who use dump on mounted filesystems. Thanks to
601 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
602 for this suggestion.
604 9. Updated the RPM spec file following the RedHat changes
605 (dynamically linked binaries now in /usr/sbin etc).
607 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
608 which enables restore to recognise multi volume compressed dumps
609 done on CD. Included his scripts for dump (which pipe the dump
610 output directly into cdrecord) and restore. There is now
611 possible to to backups to CD on the fly!
613 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
614 ===================================================================
616 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
617 Marcus Meissner (Caldera International Security Dept.).
619 2. Implement the Sun rmt extended protocol. Patch submitted
620 by Ian Gordon <iangordon@users.sourceforge.net>.
622 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
623 the linux kernel header. This ensures that dump always has the
624 latest version of this file. Patch submitted by
625 Andreas Dilger <adilger@turbolinux.com>.
627 4. Report any filesystem present in either /etc/fstab with a
628 positive passno or /etc/dumpdates in dump -w output.
629 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
631 5. Fixed the looping problem in dump introduced in the
632 previous version.
634 6. Changed the -B option of dump to limit the size of
635 _compressed_ output per volume if compression is on.
636 Patch contributed by Helmut Jarausch
637 <jarausch@igpm.rwth-aachen.de>. Note however that, since
638 it is impossible to predict the size of the compressed
639 data before doing the compression, the -B limit is a bit
640 conservative.
642 7. Fixed a bug in reading the operator typed file/tape path for
643 the next volume in restore.
645 8. Implemented a "-F script" option to restore which permits the
646 user to specify a script which will be launched at the
647 beginning of each tape, useful for automatic programming of
648 tape changers for example. See the restore man page for the
649 script parameters and return codes.
651 9. Small fix for the QFA routines provided by Uwe Gohlke
652 <uwe@ugsoft.de>, and some recommendations for QFA uses in
653 the man pages.
655 10. Fixed the multivolume restoring where making a mistake
656 to the 'Mount next tape' prompt caused several blocks to
657 be lost.
659 11. Enhanced the -e option of dump to take as a parameter a
660 comma separated list of inode numbers.
662 12. Added the -E option to dump which specify a file containing
663 inode numbers to exclude from the dump.
665 13. Fixed the compressed multi-volume dump + restore.
667 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
668 ==================================================================
670 1. Made dump report the number of blocks dumped per volume.
671 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
673 2. Fix a bug in dump where waiting too much at the 'change volumes'
674 question caused the volume to be rewritten. Thanks to
675 Graham Reed <greed@users.sourceforge.net> for reporting the
676 bug and providing a patch.
678 3. Added a compression option to dump, contributed by Jerry
679 Peters <gapeters@worldnet.att.net>.
681 WARNING: the compression works only on files/pipes and
682 drives supporting variable block size.
684 WARNING2: tapes created using the compression feature are
685 incompatible with the standard dump tape format, and a
686 version of restore >= 0.4b22 will be needed for extraction.
688 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
689 architectures. Thanks to Paul Slootman <paul@debian.org> for
690 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
691 it upstream.
693 5. Many cleanups (CPP defines, const char warnings, check of
694 ext2fs COMPAT flags, time_t cleanups, added libext2 version
695 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
697 6. Made --prefix option work in configure. All the install path
698 are now based on the configure parameters.
700 7. Added the Quick File Access mode in dump/restore, contributed
701 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
702 a file tape position for each inode, and this file is used by
703 restore (if called with parameter Q and the filename)
704 to directly position the tape at the file restore is currently
705 working on. This saves hours when restoring single files from
706 large backups, saves the tapes and the drive's head. Use
707 --enable-qfa option of configure to compile in the QFA support.
709 8. Added the possibility to dump several files and directories
710 in a single invocation of dump. Thanks to Uwe Gohlke
711 <uwe@ugsoft.de> for implementing this option.
713 9. Fixed the dumping and restoring of files with holes
714 (files ending with holes didn't get dumped or restored
715 correctly).
717 10. Fixed a socket descriptor leak in dump, which leaved opened
718 3 file descriptors per dump process (and there is one dump
719 process per tape).
721 11. Fixed dump large file system support, by explicit use of
722 open64/lseek64/etc functions (explicit use needed because
723 e2fsprogs libraries don't behave well when compiled with
726 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
727 ======================================================================
729 1. Fixed some bugs in the dump -w|-W logic introduced by
730 the previous version. Thanks to Andreas Dilger
731 <adilger@turbolinux.com> for his help on this one.
733 2. Fixed again a compilation problem when using old e2fs
734 headers (filesystem label related). Thanks to many users
735 who reported this stupid error.
737 3. Fixed a build problem on old lib5 based systems dealing with
738 _PATH_MNTTAB being undefined. Thanks to John Adams
739 <johna@onevista.com> for reporting the bug.
741 4. Improved the error detection in restore when creating
742 the temporary files in TMPDIR. Now restore will corectly
743 report a 'no space left on device' error instead of
744 strange errors which could imply an archive corruption.
745 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
746 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
748 5. Added the throughput information in dump's progression
749 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
750 for the patch.
752 6. Use libext2fs's inode scanning routines, which are particularly
753 robust when dealing with errors (especially when having some
754 bad blocks on the filesystem). Many thanks to Colin
755 <colin@horizon.com> for reporting the bug, investigating
756 the issues, finding a workaround, writing the patch and
757 fully testing it... (of course, if this patch will break
758 anything, he is to blame for :-)).
760 7. Made dump and restore LFS aware. Dump can dump a filesystem
761 containing large files, generate a large file on output and
762 restore can restore them. This feature is not enabled by
763 default, use --enable-largefile option of configure to enable
764 it (you need to have a LFS aware glibc though). Thanks to
765 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
766 and to Theodore T'so <tytso@valinux.com> for his always
767 useful thoughts.
769 8. Made dump ask upon a tape write error if it should rewrite
770 the current volume (assume this is a bad tape) or if it should
771 assume an end-of-tape condition (useful for tape drives which
772 behaves badly at the end of the tape). Thanks to Andreas
773 Dilger <adilger@turbolinux.com> for the suggestion.
775 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
776 =======================================================================
778 1. Fixed a small compilation problem due to a change
779 in the definintion of the struct sigaction in
780 glibc 2.0 / libc5. Thanks to Gunther Schlegel
781 <schlegel@riege.de> for reporting the bug and to
782 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
783 a fix.
785 2. Modified the label and uuid analysis in order to be
786 self-contained (independant of kernel/libc headers). This
787 should fix the compile with older kernel/libc headers and
788 will preserve the functionality. Thanks to Bernhard Erdmann
789 <bernhard.erdmann@gmx.de> for reporting the bug.
791 3. The 'exclude inode' option, if applied to a directory
792 inode, excludes now correctly all the leaves of this
793 directory. Thanks to John R. Dennison
794 <gerdesas@users.sourceforge.net> for reporting the bug.
796 4. Fixed the '-e' option to disable the possibility
797 to exclude the root inode (which causes the dumps to
798 be unreadable by restore). Prevented array overflow
799 when multiple -e options are used.
801 5. Fixed dump to correctly interpret a filesystem argument
802 which is a mountpoint and it is not an absolute path
803 (as specified in the fstab). Thanks to Bernhard R. Erdmann
804 <be@berdmann.de> for reporting the bug.
806 6. Made dump able to backup files larger than 2 GB. Note that
807 dump still doesn't cope with files larger than 4 GB.
809 7. Restore the real uid and gid before invoking an external
810 RSH program (big hole when dump or restore is suid root!).
812 8. Get the values for the default tape device and the location
813 of fstab file from the system headers. Thanks to
814 Andreas Dilger <adilger@turbolinux.com> for the patch.
816 9. Made dump -w|-W report all recognized filesystems
817 present in either /etc/fstab or /etc/dumpdates, and present
818 the list in the same order as in fstab file. Thanks
819 to Andreas Dilger <adilger@turbolinux.com> for the patch.
821 10. Made dump's -a (automatic end of tape detection) the
822 default. Specifying one of -B, -c, -d or -s options will
823 override the EOM detection. Thanks to Andreas Dilger
824 <adilger@turbolinux.com> for the patch.
826 11. Save the ext2 filesystem label into the dump volume label.
827 Specifying a volume label on the command line (-L option)
828 will override this feature. Thanks to Andreas Dilger
829 <adilger@turbolinux.com> for the patch.
831 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
832 =====================================================================
834 1. Fixed the signal handling in dump (which I broke in 0.4b17)
835 which was causing several strange dump failures (dump
836 hanged or segmentation faults etc).
838 2. Specified the default blocksize in the dump man page.
840 3. Changed two info messages of restore to be written on stdout
841 instead of stderr in order to leave stderr only for errors
842 or warnings. Thanks to Stephen Atwell
843 <satwell@urbana.css.mot.com> for the suggestion.
845 4. Corrected an off by one calculation which prevented
846 dumping a directory having a 1 character name length.
847 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
848 for reporting the bug.
850 5. Reinforce dump in order to allow only level 0 and no
851 -u option when dumping a subdirectory, and document
852 this in the man page. Thanks to Bernhard Erdmann
853 <bernhard.erdmann@gmx.de> for reporting the bug.
855 6. Fixed a small harmless bug in restore which caused
856 in some conditions hard links to be restored several
857 times (therefore generation some warning messages).
858 Thanks to Kenneth Porter <shiva@well.com> for
859 reporting the bug.
861 7. Updated the RPM spec file to the latest RedHat version,
862 providing FHS packaging and other cosmetic changes.
863 You will need RPM version >= 3.0.5 in order to build the RPM.
865 8. Updated the configure script to check for libtermcap
866 before checking for libreadline, because we need this
867 library in order to compile the readline support.
869 9. Made dump understand the LABEL= and UUID= notation
870 both in /etc/fstab and on the command line. Note that
871 you will need the /proc filesystem in order to use
872 these notations. Thanks to Erik Troan <ewt@redhat.com>
873 for providing the patch.
875 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
876 ===================================================================
878 1. Fixed a potential buffer overflow in restore. Thanks
879 to Stan Bubrouski <satan@fastdial.net> for reporting
880 the bug.
882 2. Fixed a readline-related bug which prevented
883 'cat DUMPFILE | restore -i -f -' from working. Thanks
884 to Charles Karney <karney@users.sourceforge.net>
885 for the bug report.
887 3. Changed a few "panic" into "exit", causing restore to
888 be more stable against some attacks (like the last one
889 reported on Bugtraq, although the last version of restore
890 was not vulnerable - just dumped core). Thanks to
891 Andreas Hasenack <andreas@conectiva.com.br> for reporting
892 the bugs to me.
894 4. Removed the suid-root bit on dump and restore in the
895 default build (and generated RPMs). It should be safer
896 now. Users who need the suid-root capabilities in order
897 to make network backups should read first the man page
898 of dump and enable the suid bit manually.
900 5. Added -ltermcap to the compile parameters for restore
901 when using readline support, in order to make the compile
902 process work on some systems (SuSE ?). Thanks to
903 Patrik Schindler <poc@pocnet.net> for reporting the bug.
905 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
906 ====================================================================
908 1. The -F script is called now *only* at the end of a tape,
909 not at the end of the dump. Thanks to Christian Haul
910 <haul@informatik.tu-darmstadt.de> for the bug report.
912 Normally, the device name and the current volume number
913 are passed on the command line of the script. However,
914 if you want the old style script (with no arguments
915 passed on the command line), you can enable it in
916 configure with the --enable-oldstylefscript.
918 2. Use posix signal handling to preserve dump functionality
919 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
920 the patch.
922 3. Made the exit code of restore in case of a 'C'ompare
923 command reflect the result of the compare. An exit status
924 of 0 means the dump archive is correct, 1 means tape errors,
925 2 means that some files were modified. Thanks to Kenneth Porter
926 <shiva@well.com> for the suggestion.
928 4. Made (finally) quotes work in interactive restore.
930 5. Small fixes in order to allow dump to compile with a
931 really old e2fsprogs version. Thanks to Ian Zimmerman
932 <itz@speakeasy.org> for the bug report.
934 6. Add GNU readline capabilities to interactive restore.
935 Use configure's --enable-readline flag to enable this feature.
936 Thanks to Patrik Schindler <poc@pocnet.net> for the
937 suggestion.
939 7. Do the compare on the fly in restore 'C' mode (this will
940 allow not to exhaust the available /tmp space when
941 ccmparing large files). Thanks to Kenneth Porter
942 <shiva@well.com> for the suggestion.
944 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
945 ====================================================================
947 1. Fixed some several duplicate 'DUMP: DUMP:' in the
948 output of dump.
950 2. Corrected the estimation of blocks to dump. Note that
951 this is still wrong for large directory inodes, due
952 to the size of a BSD directory entry on the tape
953 (estimation is based on the size of the ext2 directory,
954 which is smaller because it doesn't contain the null
955 character at the end).
957 3. Made dump report the total number of MB written to
958 tape at the end of run. Thanks to W. Reilly Cooley
959 <wcooley@nakedape.cc> for the patch.
961 4. Added the -X option to restore giving the possibility
962 to read the names of the files to be extracted/listed
963 from a text file (in addition of the command line).
964 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
965 patch.
967 5. Added the device name and the current volume number
968 as arguments to the end of tape script (dump -F option).
970 6. Made the multi-volume dump work again (after having
971 broken it in 0.4b15).
973 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
974 ===================================================================
976 1. Added a prompt command in interactive restore mode. Thanks
977 to Andreas Dilger <adilger@home.com> for the patch.
979 2. Fixed a buffer overflow problem in dump (caused by
980 not checking the size of the filesystem parameter).
981 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
982 reporting this on Bugtraq (and to several dump users
983 who forwarded me his mail).
985 3. Added the '-F script' option to dump in order to
986 launch a script at the end of each tape (to be used
987 with a tape changer, or to notify the sysadmin by
988 pager etc.).
990 4. Fixed a bug in restore compare code caused by the changes
991 I made in 0.4b14.
993 5. Fixed the treatment of options using the old BSD syntax
994 in both dump and restore.
996 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
997 =======================================================================
999 1. Fixed a bug in dump which may have caused invalid deleted
1000 directories to be dumped out if they were deleted after the
1001 mapping pass. This could occure on active filesystem and lead
1002 to heap corruption (causing dump malfunction in many possible ways).
1003 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
1004 bug and submitting the patch.
1006 2. Fixed the handling of the filesystem parameter in dump. A
1007 '/mnt/disk' parameter caused the disk contents to be dumped,
1008 but a '/mnt/disk/' parameter caused the mountpoint directory
1009 to be dumped (generally an empty directory).
1011 3. Improved the output of dump in order to tell which directory
1012 it is currently dumping (when dumping a subtree).
1014 4. Added the '-e' exclude inode option to dump. Thanks to
1015 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
1017 5. Added a REPORTING-BUGS file in order to provide a guide
1018 on how to correctly report a bug in dump/restore.
1020 6. Fixed a restore problem when restoring a hard link to an inode
1021 having the immutable or append-only attribute set. Thanks to
1022 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
1024 7. Fixed a compatibility problem between dumps made on little
1025 endian machines (the format was correct) and big endian
1026 machines (the format was incorrect). This fix break the
1027 compatibility with the older dumps made on big endian
1028 machines (sparc, m86k, ppc etc). For the first time in
1029 linux dump's history (I believe), the dumps made by *BSD,
1030 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
1031 so interchangeable. Thanks to Rob Cermak
1032 <cermak@ahab.rutgers.edu> for submitting the bug and
1033 helping me test the fix.
1035 8. Fixed the way dump reports the remaining percent/time, if
1036 the number of blocks actually dumped exceeds the estimated
1037 number of blocks. Thanks to Jean-Paul van der Jagt
1038 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
1040 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
1041 ======================================================================
1043 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
1044 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
1046 2. Removed the requirement to build the RPM as root from the
1047 spec file. Thanks to Christian Weisgerber
1048 <naddy@mips.rhein-neckar.de> for submitting this.
1050 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
1051 causing dump to block when dumping really huge filesystems.
1052 Many thanks to Patrik Schindler <poc@pocnet.net> for
1053 helping me find this bug.
1055 4. Fixed the treatment for an interrupt signal when dump access
1056 the remote tape through RSH. Thanks to Christian Weisgerber
1057 <naddy@mips.rhein-neckar.de> for providing the patch.
1059 5. Fixed a bug which was causing dump/restore to display
1060 garbage characters instead of the remote host name.
1062 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
1063 =====================================================================
1065 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
1066 <kukuk@suse.de> for submitting the patch.
1068 2. Fix for the exit code when using the size estimate option of
1069 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
1070 submitting the patch.
1072 3. Handle EINTR in atomical reads/writes in dump, which was causing
1073 dump to fail on some systems. Thanks to Eric Jergensen
1074 <eric@dvns.com> for reporting the bug and submitting the patch.
1076 4. Handle more than 16 characters for the device names in dumpdates.
1077 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
1078 tracking down the problem and proposing the solution.
1080 5. Fixed a bug in dump which prevented the creation of the
1081 dumpdates file when doing a 0-level dump without already
1082 having a dumpdates file. Thanks to Patrik Schindler
1083 <poc@pocnet.net> for reporting the bug.
1085 6. Changed the way dump 'S' flag reports the size estimate
1086 from number of blocks into bytes (making it compatible
1087 with the Solaris version, and simplifying things for
1088 amanda users). Thanks to Jason L Tibbitts III
1089 <tibbs@math.uh.edu> for reporting the bug.
1091 7. Fixed a compatibility problem in linux/alpha dump tape format.
1092 Now the linux/alpha dump are (again) compatible with the
1093 other dump formats. But this breaks compatibility with
1094 older dumps made on alpha. Thanks to Mike Tibor
1095 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
1097 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
1098 ======================================================================
1100 1. Added a '--enable-kerberos' to configure.
1102 2. Added a 'S' option to dump which determines the amount of space
1103 that is needed to perform the dump without actually doing it, similar
1104 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
1105 <cermak@ahab.rutgers.edu>.
1107 3. Added a 'M' multi-volume option to dump and restore which enables
1108 dumping to multiple files (useful when dumping to an ext2
1109 partition to make several dump files in order to bypass the 2GB
1110 file size limitation). The 'f' argument is treated as a prefix and
1111 the output files will be named <prefix>001, <prefix>002 etc. With
1112 the 'M' flag, restore automatically selects the right file without
1113 asking to enter a new tape each time.
1115 4. Fixed a memory leak which was causing dump to grow very big
1116 (270MB when dumping a 10GB filesystem...). Thanks to Jason
1117 Fearon <jasonf@netrider.org.au> for reporting the bug.
1119 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
1120 ======================================================================
1122 1. Make configure test if the system glob routines support
1123 extended syntax (ALTDIRFUNC). If not, use the internal glob
1124 routines instead of system ones. Thanks to Bernhard Sadlowski
1125 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
1126 and helping me resolve this and other minor libc5 compiling
1127 glitches.
1129 2. Fix a problem when dumping a ext2fs with the 'filetype'
1130 feature enabled. Thanks to Patrick J. LoPresti
1131 <patl@cag.lcs.mit.edu> for reporting the bug and to
1132 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
1134 3. Made the nodump flag work on directories. A directory which
1135 has the nodump flag gets never dumped, regardless of its
1136 contents.
1138 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
1139 which allows dump on an active ext3 filesystem. However, this
1140 is a "quick and dirty" patch which enables backup of an ext3
1141 filesystem through the ext2 compatibility (by ignoring the
1142 NEEDS_RECOVERY bit). The journal file is not recognized and
1143 it is dumped (it should not).
1145 5. Test the superblock compatibility flags when dumping, in order
1146 to be sure that we know how to deal with specific features.
1148 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
1149 ====================================================================
1151 1. Use lchown instead of chown, fixing a possible security problem
1152 when restoring symlinks (a malicious user could use this
1153 to deliberately corrupt the ownership of important system files).
1154 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
1155 this and providing the patch.
1157 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
1158 ====================================================================
1160 1. Put dump sources under CVS, added Id tags in all files so
1161 one can use 'ident' on binary files.
1163 2. Added the dump/restore version in the usage text so one can
1164 easily verify the version he is using.
1166 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
1167 a va_start/va_end problem on linux-ppc (always call va_start
1168 va_end in pairs each time we use a vararg function).
1170 4. Added again the DT_* constants because old libc does not
1171 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
1172 for submitting the bug report.
1174 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
1175 this should enable dumping big (huge) filesystems.
1177 6. Added the RSH environment variable in order to be able to
1178 use a rsh replacement like ssh when doing remote backups (and
1179 bypass the security limitations of rcmd). Now you can do remote
1180 backups without being root (or making dump setuid root).
1182 7. Modified again the way dumpdates works. For incremental dumps,
1183 we need to read dumpdates even if we are not using 'u' option.
1184 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
1185 this should work.
1187 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
1188 ===================================================================
1190 1. Removed the 'k' flag from the restore 'about' text if kerberos
1191 was not compiled in.
1193 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
1194 to them (fsetflags takes a char*, setflags an open fd!).
1196 3. (f)setflags is called only if the flags aren't empty. If the
1197 file is a special file, a warning is printed, because changing
1198 flags implies opening the device. Normally, a special file
1199 should not have any flag... (Debian bug #29775, patch provided
1200 by Abhijit Dasgupta <abhijit@ans.net>).
1202 4. Made possible to dump a file system not mentioned in /etc/fstab.
1203 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
1205 5. Changed the default behaviour to not create dumpdates
1206 unless 'u' option is specified. Removed the old "debian-patch"
1207 which provided the same thing. (Debian bug #38136, #33818).
1209 6. Removed all those dump*announce, since they were getting old...
1211 7. Added warning messages if dumpdates does not exist and
1212 when an inferior level dump does not exist (except for a level 0
1213 dump).
1215 8. Debugged the glob calls in interactive mode: restore used a
1216 dirent struct which was different from the /usr/include/dirent.h
1217 one (this used to work, is it a glibc2 change?), so none of the
1218 compat glob (which used /usr/include/dirent.h) or the system glob
1219 worked. Restore use now the system dirent (and the system
1220 DT_* constants), which are compatible with BSD ones.
1222 9. Added a configure flag (--with-dumpdatespath) to specify
1223 the location of dumpdates file. By default, it is
1224 /etc/dumpdates.
1226 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1227 included the current date/version in man pages.
1229 11. Corrected the estimation of remaining time when
1230 the operator doesn't change the tapes quickly enough. This
1231 was an old bug, I thought I corrected it, and discovered
1232 that in fact it was corrected in two different places, so
1233 the results canceled each other...
1235 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1236 ===================================================================
1238 1. Integrated multiple patches from RedHat, Debian and SuSE:
1240 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1241 isn't specified.
1242 - several fixes in the man pages.
1243 - update the default tape device to /dev/st0.
1244 - many updates for Linux Alpha (byte ordering, size_t etc).
1245 - buffer overruns.
1246 - use environment variable for TMPDIR (instead of /tmp).
1247 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1248 - workaround egcs bug (RedHat bugs #4281 and #2989).
1249 - wire $(OPT) throughout Makefile's.
1251 2. Upgrade the dump revision to 1, making possible to dump filesystems
1252 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1254 3. Fix some compile warnings, prototype all functions.
1256 4. Use glibc err/glob instead of internal compatibility
1257 routines (only if available).
1259 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1260 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1262 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1263 ======================================================================
1265 1. Integrated the changes from FreeBSD-3.1-RELEASE
1266 (mostly bug fixes, buffer overruns, dump has now an "automatic
1267 tape length calculation" flag, dump/restore can use kerberos now
1268 (this is NOT tested), use environment variables for TAPE and
1269 RMT etc.).
1271 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1273 3. Corrected a bug in restore when using 'C' option with multi-volumes
1274 tapes (files splited accros two tapes give "size changed" errors
1275 when comparing).
1277 4. Corrected the long standing bug when dumping multiple tapes.
1278 This works for me, needs further testing.
1280 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1281 ====================================================================
1283 1. Dump now runs correctly on kernels 2.1.x
1284 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1286 Changes between versions 0.4b2 and 0.4b3
1287 ========================================
1289 1. Use realpath() if available
1291 2. Report statistics
1293 Changes between versions 0.4b1 and 0.4b2
1294 ========================================
1296 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1297 the patch)
1299 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1301 3. Fixed configure.in to correctly handle --enable-rmt
1303 Changes between versions 0.3 and 0.4b1
1304 ======================================
1306 1. Integrated the changes from 4.4BSD-Lite2
1308 2. Integrated the patches from Debian and Red Hat
1310 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1312 4. Changed dump to use the Ext2fs library to get block addresses. This
1313 should solve the endianness problem on SparcLinux.
1315 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1316 distribution's one) to use autoconf
1318 6. Fixed a few minor bugs
1320 Changes between versions 0.2e and 0.2f
1321 ======================================
1323 1. Added the creation of named pipes (fifos) in restore.
1325 2. Added the -N flag in the restore manual page.
1327 3. Added the file kernel.patch which contains the llseek() optimization
1328 patch for 1.2.x kernels.
1330 4. Fixed a bug in the restoration of symbolic links: owner and group were
1331 not restored.
1333 5. Integrated some changes from FreeBSD 2.2.
1335 6. Added a call to ftruncate() after restoring each file to restore
1336 correctly files ending by a hole.
1338 Changes between versions 0.2d and 0.2e
1339 ======================================
1341 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1342 with owner = root (0) and group = root/wheel/whatever (0).
1344 Changes between versions 0.2c and 0.2d
1345 ======================================
1347 1. Dump is now able to backup 2GB+ filesystems.
1349 2. Dump and restore can now be linked as static binaries.
1351 Changes between versions 0.2b and 0.2c
1352 ======================================
1354 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1356 Changes between versions 0.2a and 0.2b
1357 ======================================
1359 1. Really fixed the bug that I should have corrected in 0.2a.
1361 2. Enabled optimization again.
1363 Changes between versions 0.2 and 0.2a
1364 =====================================
1366 1. Disabled the optimization during compilation.
1368 Changes between versions 0.1 and 0.2
1369 ====================================
1371 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1372 the fs_type field (actually, I modified the file fstab.c to make it
1373 use the mntent functions).
1375 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1376 made by dump should be readable by the BSD restore and backups made
1377 by the BSD dump should be readable by restore. Unfortunately, this
1378 means that the dump format has changed between version 0.1 and version
1379 0.2 :-(
1381 3. Dump is now able to backup a subtree, it is no longer limited to whole
1382 filesystems like the BSD version.
1384 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1385 than 2 GB.
1387 Changes between versions 0.0 and 0.1
1388 ====================================
1390 1. Now create links rdump and rrestore during the `make install' step.
1392 2. Linux port specific bugs added to the manual pages
1394 3. Incorrect estimation of the number of tapes blocks fixed when doing
1395 incremental backups.
1397 4. Better ls-like format in restore in interactive mode.