1 $Id: CHANGES,v 1.212 2003/02/11 08:57:41 stelian Exp $
3 Changes between versions 0.4b33 and 0.4b34 (released ?????????????????)
4 =======================================================================
6 1. Fixed the INSTALL file to reflect the actual install paths.
7 Thanks to David Ronis <ronis@ronispc.chem.mcgill.ca> for
10 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
11 =======================================================================
13 1. Added a note in the restore man page clarifying the question
14 'set the permissions on the current directory ?' asked by
15 restore at the end of treatment in -i and -x modes.
17 2. Fixed the endianess issues when reading compressed tapes.
18 Thanks to Dark Force <daq4th@users.sourceforge.net> for
19 reporting this bug and providing test cases.
21 3. Fixed the "ACL won't be dumped" warning message (which showed
22 an extra, unrelated error message). Thanks to Dragan Krnic
23 <dkrnic@lycos.com> for reporting this bug.
25 4. Made dump look first into /etc/mtab, then into /etc/fstab
26 when searching for filesystem paths. Also fixed some problems
27 caused by binding mounts or multiple block device mounts.
28 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
29 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
30 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
31 and #160305, Sourceforge bugs #618699 and #679832).
33 5. Made dump's -I option accept the value '0' meaning all the
34 read errors will be ignored. This can be useful when running
35 dump from unattended sessions (like cron jobs). Thanks to
36 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
38 6. Fixed the output of dump to indicate 'blocks' instead of
39 'tape blocks' in the various messages (blocks are always
40 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
41 made it clearly print the current blocksize at the start of
42 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
45 7. Made rmt understand long (largefiles) seeks.
47 8. Fixed build with very old versions of libext2fs, where
48 EXT2_FT_* constants were undefined.
50 9. Made dump accept the dumpdates path on the command line
51 (-D file option) instead of using only the hardcoded one.
52 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
55 10. Enabled rmt, LFS, readline, QFA options by default in
56 ./configure. Updated the configure process (new versions
57 of config.guess, config.sub etc).
59 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
60 =======================================================================
62 1. Changed dump to use fcntl(F_SETLK) style locking instead
63 of flock() when locking the dumpdates file. With the old
64 locking scheme, a local user having read rights on the
65 dumpdates file could be able to do a Denial of Service attack
66 on dump. In order to lock the dumpdates file with the new
67 scheme, the user would need to have write access on the file.
68 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
69 reporting the bug (originally a bugtraq post).
71 2. Fixed interactive 'ls' which caused spurious errors warnings
72 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
73 <jorgen_ostling@users.sourceforge.net> for reporting this
76 3. Fixed dump's estimate when dealing with sparse inodes.
78 4. Modified dump to allow setting a blocksize bigger than 32kB
79 (raised the limit to 1024kB), because newer hardware needs
80 this for proper operation. Thanks to Dirk Traenapp
81 <dtraenapp@users.sourceforge.net> for reporting this.
83 5. Fixed a bug causing Dump to stop and report an error if an
84 inode number in the exclude file was followed by some amount
85 of whitespace. Thanks to Jeffrey Sofferin
86 <sofferin@users.sourceforge.net> for reporting this bug.
88 6. Fixed a bug which caused restore, in some particular cases,
89 to ask some 'scary' questions and leave a bunch of RSTTMP
90 directories behind when restoring incremental tapes. Thanks
91 to Philippe Troin <phil@fifi.org> for reporting this bug and
92 providing the test cases.
94 7. Changed the wording when inodes are excluded from dump:
95 replaced 'Added inode 7 to exclude list' with
96 'Excluding inode 7 (resize inode) from dump', as suggested
97 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
99 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
100 ===================================================================
102 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
103 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
104 <eros@lamel.bo.cnr.it> for reporting the bug.
106 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
107 ===================================================================
109 1. Made dump print out the ext2 logical block number in case of a read
110 error, which can be given as an argument to debugfs' ncheck command,
111 in order to find out the affected inode. Added note about this
112 usage in dump's man page.
114 2. Fixed a problem in restore when reading tapes written on big
115 endian machines with very old versions of dump. The patch was
116 contributed by George Helffrich <george@geology.bristol.ac.uk>.
118 3. Fixed the tape length calculation when using large tapes
119 and compression. Thanks to Georg Lippold
120 <g_lippold@sourceforge.net> for reporting the bug.
122 4. Added a new set of examples in dump_on_cd_2 directory, based
123 on dump_on_cd examples but somewhat enhanced, supporting DVD
124 media, and localized in english and german. Thanks to
125 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
127 5. Made dump save 32 bit UID/GID. Internally, this was achieved
128 by switching from the old BSD inode format to the new BSD
129 inode format, which means that the tape format was changed.
130 However, since all restore versions out there should
131 transparently support both inode formats, the change should
132 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
135 6. Fixed a lot of warnings in the code shown when compiling
136 with 'gcc -W'. Thanks to Matthias Andree
137 <matthias.andree@stud.uni-dortmund.de> for reporting this.
139 7. Fixed a small markup bug in the dump man page. Thanks to
140 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
143 8. Rewrote entirely the man pages using the tmac.an macro
144 package (Linux man page format) instead of the original BSD
145 format. They should be now cleaner and easier to modify.
147 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
148 ==================================================================
150 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
151 non Linux clients were misinterpreted. The description of the problem
152 (incompatible numbering in Linux mtio opcodes) is documented at
153 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
154 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
155 providing an excellent, cross-platform replacement for rmt in his
158 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
159 which caused restore to fail when the inode maps were not entirely
160 contained on the first volume (when using really small volumes or
161 when dumping a huge number of inodes).
163 3. Fixed a problem in dump, where files in subdirectories of directories
164 that have the nodump flag set (or being excluded with -e/-E) were
165 sometimes incorrectly dumped (instead of being excluded). The original
166 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
167 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
168 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
169 help testing the patch.
171 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
172 which makes one able to pipe the output of dump, by the net, to
173 a remote CD-burner server.
175 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
176 and use GNU's symbolic syntax over rmt instead of numerical values
177 to assure multiple platform compatibility.
179 6. Documented the -d option in restore.
181 7. Added a -v (verbose) mode to dump. For now it just prints the number
182 of the inode being dumped, but this could evolve in future versions
183 to include interesting debugging output.
185 8. Added a -o flag to restore, which automatically restores the current
186 directory permissions in -i or -x mode, without asking the operator
187 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
188 and Peter Samuel <peters@e-smith.com>.
190 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
191 ====================================================================
193 1. Fixed a bug in the QFA feature which could have caused
194 a bad tape positionning on restore (causing extra delays in
195 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
197 2. Added a small note in the dump man page specifying that
198 there should be no white space between the option letter and
199 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
201 3. Made restore work with older versions of the readline library,
202 by checking in configure for several extended symbols. Restore
203 can now be compiled with a readline as old as the 2.0 release
204 (though it may be a good idea to upgrade it to a more recent
205 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
206 reporting the build failures.
208 4. Fixed a performance problem with the QFA file creation in
209 dump, which made unnecessary seeks on the tape slowing down
210 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
211 reporting this issue.
213 5. Removed the inclusion of some kernel headers in the dump
214 source, which prevented the compile in some kernel/glibc
215 headers/architecture combination. Thanks to Bdale Garbee
216 <bdale@gag.com> for reporting the bug.
218 6. Added the appropriate error message when dump fails to
219 open the output file for writing. Thanks to Amith Varghese
220 <amithv@yahoo.com> for reporting this bug.
222 7. Made restore able to understand large Solaris ufsdump tapes
223 (containing inodes bigger than 4194304). Sun have introduced
224 an "extension" to the dump tape format when dealing with
225 those inodes, which was uncorrectly handled by Linux restore.
226 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
227 providing a test case.
229 8. Added the -m parameter to dump which optimises the output for
230 inodes having been changed but not modified since the last dump
231 ('changed' and 'modified' have the meaning defined in stat(2)).
232 For those inodes, dump will save only the metadata, instead of
233 saving the entire inode contents. Inodes which are either
234 directories or have been modified since the last dump are saved
235 in a regular way. Uses of this flag must be consistent, meaning
236 that either every dump in an incremental dump set have the flag,
239 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
240 =======================================================================
242 1. Fixed behaviour of dump when exceeding resource limits
245 2. Added the -L flag to restore to allow the user to specify a
246 maximal allowed number of miscompares when using restore
247 with the -C option to check the backup.
249 3. Detailed the manual entry for the -N option of restore.
251 4. Added the -a flag to restore to make able doing unattended
252 restores in -i or -x mode (automatically walks through the
253 multiple dump volumes).
255 5. Extended the QFA mode to work with local files and/or
256 remote tapes and files. This way, restore can know in advance
257 the tape number and the offset for the inodes to extract and
258 can minimize the extraction time by seeking directly to the
261 6. Added the -A <archive> option to both dump and restore,
262 which makes dump to archive a dump table-of-contents in
263 the specified file to be used by restore to determine
264 whether a file is in the dump file that is being restored.
265 (the archive file syntax is also compatible with the
266 Solaris ufsdump generated one).
268 7. Small fix in restore making it able to read some (broken ?)
269 Solaris ufsdump tapes.
271 8. Fixed dump to correctly recognise the root filesystem when using
272 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
273 for reporting this bug.
275 9. Added the -P <file> option to restore to create a
276 Quick File Access file from an already made dump. Patch
277 contributed by Uwe Gohlke <uwe@ugsoft.de>.
279 10. Made restore compile and run on Solaris, making it a
280 possible replacement for the standard ufsrestore. Port was
281 contributed by Uwe Gohlke <uwe@ugsoft.de>.
283 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
284 =====================================================================
286 1. Added a set of backup scripts from Eugenio Diaz
287 <getnito@yahoo.com> in the examples section. It features
288 automatic (cron based) full and incremental dumping of
289 several filesystems on a separate filesystem.
291 2. Fixed a off-by-one miscalculation which disabled dumping
292 a one letter subdirectory. Thanks to Chris Siebenmann
293 <cks@utcc.utoronto.ca> for reporting the bug.
295 3. Fixed several restore bugs occuring when trying to
296 restore some missing files on the tape. Thanks to Chris
297 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
299 4. Fixed --with-ldopts configure argument passing, installing from
300 a separate object directory, makefile cleanups contributed
301 by <splite@purdue.edu>.
303 5. Fix a bug which could caused, in some conditions, the highest
304 number inode of a filesystem, to not be dumped. Many thanks
305 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
308 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
309 =======================================================================
311 1. Added a mini howto from Patrick Walsh in the examples
314 2. Minor man pages syntax corrections. Thanks to
315 Chris Danis <screechco@home.com> for reporting the bugs.
317 3. Added a script from David B. Peterson <dave@toppledwagon.com>
318 to the examples section. It features dumping several
319 filesystems to a remote tape drive upon ssh.
321 4. Added a patch provided by Richard Jones <rich@annexia.org>
322 which allows BRADEMAX (number of read errors tolerated by
323 dump) to be adjusted using the -I option.
325 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
326 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
329 6. Add the -l option to restore to specify if, when doing a
330 remote restore, the file used to access the backup is a
331 regular file (the defaults being a tape drive). Restore needs
332 to know this information when reading a remote compressed
333 dump. Previously, this information was autodetected, but
334 the autodetection code fails (with ioctl: Inappropriate ioctl
335 for device) when using a non Linux remote box. Thanks to
336 many users and especially to Eros Albertazzi
337 <eros@lamel.bo.cnr.it> for reporting this.
339 7. Found a workaroung for the dump deadlock problem (3 childs
340 stuck in pause(), father in read()). The workaround seems
341 to work for me and several beta-testers. If it doesn't work
342 for you, please report back.
344 8. Updated the RPM spec file (BuildPrereq, URL etc).
346 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
347 ========================================================================
349 1. Fixed the permissions of a newly created QFA file by dump.
351 2. Cleaned up the source of dump (the external variables
352 definition was a complete mess, making possible to have
355 3. Fixed restore to use the full tape volume path when doing
356 a compare (since it changes the working directory to the
357 filesystem being compared in the process).
359 4. Added the -q option to dump which makes dump abort
360 whenever it needs operator attention. It should be
361 easier to use dump in scripts with this option.
363 5. Detect the use of incompatible options to dump and
364 refuse them (like -a and -B options together).
366 6. Added bzip2 compression to dump/restore (use option -j level
367 to select it). Note that you will need a restore version
368 >= 0.4b24 in order to restore a bzip2 compressed dump.
369 The same warning as for the zlib compression applies:
370 the tape format of a bzip2 dump is not compatible with the
371 original BSD tape format.
373 7. Fixed a overflow problem in dump corrupting the dump when
374 very large files were encountered. Thanks to Vladimir Nadvornik
377 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
378 kernel buffer/page cache before starting the dump, helping
379 a bit those who use dump on mounted filesystems. Thanks to
380 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
383 9. Updated the RPM spec file following the RedHat changes
384 (dynamically linked binaries now in /usr/sbin etc).
386 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
387 which enables restore to recognise multi volume compressed dumps
388 done on CD. Included his scripts for dump (which pipe the dump
389 output directly into cdrecord) and restore. There is now
390 possible to to backups to CD on the fly!
392 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
393 ===================================================================
395 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
396 Marcus Meissner (Caldera International Security Dept.).
398 2. Implement the Sun rmt extended protocol. Patch submitted
399 by Ian Gordon <iangordon@users.sourceforge.net>.
401 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
402 the linux kernel header. This ensures that dump always has the
403 latest version of this file. Patch submitted by
404 Andreas Dilger <adilger@turbolinux.com>.
406 4. Report any filesystem present in either /etc/fstab with a
407 positive passno or /etc/dumpdates in dump -w output.
408 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
410 5. Fixed the looping problem in dump introduced in the
413 6. Changed the -B option of dump to limit the size of
414 _compressed_ output per volume if compression is on.
415 Patch contributed by Helmut Jarausch
416 <jarausch@igpm.rwth-aachen.de>. Note however that, since
417 it is impossible to predict the size of the compressed
418 data before doing the compression, the -B limit is a bit
421 7. Fixed a bug in reading the operator typed file/tape path for
422 the next volume in restore.
424 8. Implemented a "-F script" option to restore which permits the
425 user to specify a script which will be launched at the
426 beginning of each tape, useful for automatic programming of
427 tape changers for example. See the restore man page for the
428 script parameters and return codes.
430 9. Small fix for the QFA routines provided by Uwe Gohlke
431 <uwe@ugsoft.de>, and some recommendations for QFA uses in
434 10. Fixed the multivolume restoring where making a mistake
435 to the 'Mount next tape' prompt caused several blocks to
438 11. Enhanced the -e option of dump to take as a parameter a
439 comma separated list of inode numbers.
441 12. Added the -E option to dump which specify a file containing
442 inode numbers to exclude from the dump.
444 13. Fixed the compressed multi-volume dump + restore.
446 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
447 ==================================================================
449 1. Made dump report the number of blocks dumped per volume.
450 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
452 2. Fix a bug in dump where waiting too much at the 'change volumes'
453 question caused the volume to be rewritten. Thanks to
454 Graham Reed <greed@users.sourceforge.net> for reporting the
455 bug and providing a patch.
457 3. Added a compression option to dump, contributed by Jerry
458 Peters <gapeters@worldnet.att.net>.
460 WARNING: the compression works only on files/pipes and
461 drives supporting variable block size.
463 WARNING2: tapes created using the compression feature are
464 incompatible with the standard dump tape format, and a
465 version of restore >= 0.4b22 will be needed for extraction.
467 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
468 architectures. Thanks to Paul Slootman <paul@debian.org> for
469 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
472 5. Many cleanups (CPP defines, const char warnings, check of
473 ext2fs COMPAT flags, time_t cleanups, added libext2 version
474 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
476 6. Made --prefix option work in configure. All the install path
477 are now based on the configure parameters.
479 7. Added the Quick File Access mode in dump/restore, contributed
480 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
481 a file tape position for each inode, and this file is used by
482 restore (if called with parameter Q and the filename)
483 to directly position the tape at the file restore is currently
484 working on. This saves hours when restoring single files from
485 large backups, saves the tapes and the drive's head. Use
486 --enable-qfa option of configure to compile in the QFA support.
488 8. Added the possibility to dump several files and directories
489 in a single invocation of dump. Thanks to Uwe Gohlke
490 <uwe@ugsoft.de> for implementing this option.
492 9. Fixed the dumping and restoring of files with holes
493 (files ending with holes didn't get dumped or restored
496 10. Fixed a socket descriptor leak in dump, which leaved opened
497 3 file descriptors per dump process (and there is one dump
500 11. Fixed dump large file system support, by explicit use of
501 open64/lseek64/etc functions (explicit use needed because
502 e2fsprogs libraries don't behave well when compiled with
503 FILE_OFFSET_BITS=64).
505 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
506 ======================================================================
508 1. Fixed some bugs in the dump -w|-W logic introduced by
509 the previous version. Thanks to Andreas Dilger
510 <adilger@turbolinux.com> for his help on this one.
512 2. Fixed again a compilation problem when using old e2fs
513 headers (filesystem label related). Thanks to many users
514 who reported this stupid error.
516 3. Fixed a build problem on old lib5 based systems dealing with
517 _PATH_MNTTAB being undefined. Thanks to John Adams
518 <johna@onevista.com> for reporting the bug.
520 4. Improved the error detection in restore when creating
521 the temporary files in TMPDIR. Now restore will corectly
522 report a 'no space left on device' error instead of
523 strange errors which could imply an archive corruption.
524 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
525 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
527 5. Added the throughput information in dump's progression
528 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
531 6. Use libext2fs's inode scanning routines, which are particularly
532 robust when dealing with errors (especially when having some
533 bad blocks on the filesystem). Many thanks to Colin
534 <colin@horizon.com> for reporting the bug, investigating
535 the issues, finding a workaround, writing the patch and
536 fully testing it... (of course, if this patch will break
537 anything, he is to blame for :-)).
539 7. Made dump and restore LFS aware. Dump can dump a filesystem
540 containing large files, generate a large file on output and
541 restore can restore them. This feature is not enabled by
542 default, use --enable-largefile option of configure to enable
543 it (you need to have a LFS aware glibc though). Thanks to
544 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
545 and to Theodore T'so <tytso@valinux.com> for his always
548 8. Made dump ask upon a tape write error if it should rewrite
549 the current volume (assume this is a bad tape) or if it should
550 assume an end-of-tape condition (useful for tape drives which
551 behaves badly at the end of the tape). Thanks to Andreas
552 Dilger <adilger@turbolinux.com> for the suggestion.
554 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
555 =======================================================================
557 1. Fixed a small compilation problem due to a change
558 in the definintion of the struct sigaction in
559 glibc 2.0 / libc5. Thanks to Gunther Schlegel
560 <schlegel@riege.de> for reporting the bug and to
561 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
564 2. Modified the label and uuid analysis in order to be
565 self-contained (independant of kernel/libc headers). This
566 should fix the compile with older kernel/libc headers and
567 will preserve the functionality. Thanks to Bernhard Erdmann
568 <bernhard.erdmann@gmx.de> for reporting the bug.
570 3. The 'exclude inode' option, if applied to a directory
571 inode, excludes now correctly all the leaves of this
572 directory. Thanks to John R. Dennison
573 <gerdesas@users.sourceforge.net> for reporting the bug.
575 4. Fixed the '-e' option to disable the possibility
576 to exclude the root inode (which causes the dumps to
577 be unreadable by restore). Prevented array overflow
578 when multiple -e options are used.
580 5. Fixed dump to correctly interpret a filesystem argument
581 which is a mountpoint and it is not an absolute path
582 (as specified in the fstab). Thanks to Bernhard R. Erdmann
583 <be@berdmann.de> for reporting the bug.
585 6. Made dump able to backup files larger than 2 GB. Note that
586 dump still doesn't cope with files larger than 4 GB.
588 7. Restore the real uid and gid before invoking an external
589 RSH program (big hole when dump or restore is suid root!).
591 8. Get the values for the default tape device and the location
592 of fstab file from the system headers. Thanks to
593 Andreas Dilger <adilger@turbolinux.com> for the patch.
595 9. Made dump -w|-W report all recognized filesystems
596 present in either /etc/fstab or /etc/dumpdates, and present
597 the list in the same order as in fstab file. Thanks
598 to Andreas Dilger <adilger@turbolinux.com> for the patch.
600 10. Made dump's -a (automatic end of tape detection) the
601 default. Specifying one of -B, -c, -d or -s options will
602 override the EOM detection. Thanks to Andreas Dilger
603 <adilger@turbolinux.com> for the patch.
605 11. Save the ext2 filesystem label into the dump volume label.
606 Specifying a volume label on the command line (-L option)
607 will override this feature. Thanks to Andreas Dilger
608 <adilger@turbolinux.com> for the patch.
610 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
611 =====================================================================
613 1. Fixed the signal handling in dump (which I broke in 0.4b17)
614 which was causing several strange dump failures (dump
615 hanged or segmentation faults etc).
617 2. Specified the default blocksize in the dump man page.
619 3. Changed two info messages of restore to be written on stdout
620 instead of stderr in order to leave stderr only for errors
621 or warnings. Thanks to Stephen Atwell
622 <satwell@urbana.css.mot.com> for the suggestion.
624 4. Corrected an off by one calculation which prevented
625 dumping a directory having a 1 character name length.
626 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
627 for reporting the bug.
629 5. Reinforce dump in order to allow only level 0 and no
630 -u option when dumping a subdirectory, and document
631 this in the man page. Thanks to Bernhard Erdmann
632 <bernhard.erdmann@gmx.de> for reporting the bug.
634 6. Fixed a small harmless bug in restore which caused
635 in some conditions hard links to be restored several
636 times (therefore generation some warning messages).
637 Thanks to Kenneth Porter <shiva@well.com> for
640 7. Updated the RPM spec file to the latest RedHat version,
641 providing FHS packaging and other cosmetic changes.
642 You will need RPM version >= 3.0.5 in order to build the RPM.
644 8. Updated the configure script to check for libtermcap
645 before checking for libreadline, because we need this
646 library in order to compile the readline support.
648 9. Made dump understand the LABEL= and UUID= notation
649 both in /etc/fstab and on the command line. Note that
650 you will need the /proc filesystem in order to use
651 these notations. Thanks to Erik Troan <ewt@redhat.com>
652 for providing the patch.
654 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
655 ===================================================================
657 1. Fixed a potential buffer overflow in restore. Thanks
658 to Stan Bubrouski <satan@fastdial.net> for reporting
661 2. Fixed a readline-related bug which prevented
662 'cat DUMPFILE | restore -i -f -' from working. Thanks
663 to Charles Karney <karney@users.sourceforge.net>
666 3. Changed a few "panic" into "exit", causing restore to
667 be more stable against some attacks (like the last one
668 reported on Bugtraq, although the last version of restore
669 was not vulnerable - just dumped core). Thanks to
670 Andreas Hasenack <andreas@conectiva.com.br> for reporting
673 4. Removed the suid-root bit on dump and restore in the
674 default build (and generated RPMs). It should be safer
675 now. Users who need the suid-root capabilities in order
676 to make network backups should read first the man page
677 of dump and enable the suid bit manually.
679 5. Added -ltermcap to the compile parameters for restore
680 when using readline support, in order to make the compile
681 process work on some systems (SuSE ?). Thanks to
682 Patrik Schindler <poc@pocnet.net> for reporting the bug.
684 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
685 ====================================================================
687 1. The -F script is called now *only* at the end of a tape,
688 not at the end of the dump. Thanks to Christian Haul
689 <haul@informatik.tu-darmstadt.de> for the bug report.
691 Normally, the device name and the current volume number
692 are passed on the command line of the script. However,
693 if you want the old style script (with no arguments
694 passed on the command line), you can enable it in
695 configure with the --enable-oldstylefscript.
697 2. Use posix signal handling to preserve dump functionality
698 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
701 3. Made the exit code of restore in case of a 'C'ompare
702 command reflect the result of the compare. An exit status
703 of 0 means the dump archive is correct, 1 means tape errors,
704 2 means that some files were modified. Thanks to Kenneth Porter
705 <shiva@well.com> for the suggestion.
707 4. Made (finally) quotes work in interactive restore.
709 5. Small fixes in order to allow dump to compile with a
710 really old e2fsprogs version. Thanks to Ian Zimmerman
711 <itz@speakeasy.org> for the bug report.
713 6. Add GNU readline capabilities to interactive restore.
714 Use configure's --enable-readline flag to enable this feature.
715 Thanks to Patrik Schindler <poc@pocnet.net> for the
718 7. Do the compare on the fly in restore 'C' mode (this will
719 allow not to exhaust the available /tmp space when
720 ccmparing large files). Thanks to Kenneth Porter
721 <shiva@well.com> for the suggestion.
723 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
724 ====================================================================
726 1. Fixed some several duplicate 'DUMP: DUMP:' in the
729 2. Corrected the estimation of blocks to dump. Note that
730 this is still wrong for large directory inodes, due
731 to the size of a BSD directory entry on the tape
732 (estimation is based on the size of the ext2 directory,
733 which is smaller because it doesn't contain the null
734 character at the end).
736 3. Made dump report the total number of MB written to
737 tape at the end of run. Thanks to W. Reilly Cooley
738 <wcooley@nakedape.cc> for the patch.
740 4. Added the -X option to restore giving the possibility
741 to read the names of the files to be extracted/listed
742 from a text file (in addition of the command line).
743 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
746 5. Added the device name and the current volume number
747 as arguments to the end of tape script (dump -F option).
749 6. Made the multi-volume dump work again (after having
750 broken it in 0.4b15).
752 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
753 ===================================================================
755 1. Added a prompt command in interactive restore mode. Thanks
756 to Andreas Dilger <adilger@home.com> for the patch.
758 2. Fixed a buffer overflow problem in dump (caused by
759 not checking the size of the filesystem parameter).
760 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
761 reporting this on Bugtraq (and to several dump users
762 who forwarded me his mail).
764 3. Added the '-F script' option to dump in order to
765 launch a script at the end of each tape (to be used
766 with a tape changer, or to notify the sysadmin by
769 4. Fixed a bug in restore compare code caused by the changes
772 5. Fixed the treatment of options using the old BSD syntax
773 in both dump and restore.
775 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
776 =======================================================================
778 1. Fixed a bug in dump which may have caused invalid deleted
779 directories to be dumped out if they were deleted after the
780 mapping pass. This could occure on active filesystem and lead
781 to heap corruption (causing dump malfunction in many possible ways).
782 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
783 bug and submitting the patch.
785 2. Fixed the handling of the filesystem parameter in dump. A
786 '/mnt/disk' parameter caused the disk contents to be dumped,
787 but a '/mnt/disk/' parameter caused the mountpoint directory
788 to be dumped (generally an empty directory).
790 3. Improved the output of dump in order to tell which directory
791 it is currently dumping (when dumping a subtree).
793 4. Added the '-e' exclude inode option to dump. Thanks to
794 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
796 5. Added a REPORTING-BUGS file in order to provide a guide
797 on how to correctly report a bug in dump/restore.
799 6. Fixed a restore problem when restoring a hard link to an inode
800 having the immutable or append-only attribute set. Thanks to
801 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
803 7. Fixed a compatibility problem between dumps made on little
804 endian machines (the format was correct) and big endian
805 machines (the format was incorrect). This fix break the
806 compatibility with the older dumps made on big endian
807 machines (sparc, m86k, ppc etc). For the first time in
808 linux dump's history (I believe), the dumps made by *BSD,
809 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
810 so interchangeable. Thanks to Rob Cermak
811 <cermak@ahab.rutgers.edu> for submitting the bug and
812 helping me test the fix.
814 8. Fixed the way dump reports the remaining percent/time, if
815 the number of blocks actually dumped exceeds the estimated
816 number of blocks. Thanks to Jean-Paul van der Jagt
817 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
819 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
820 ======================================================================
822 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
823 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
825 2. Removed the requirement to build the RPM as root from the
826 spec file. Thanks to Christian Weisgerber
827 <naddy@mips.rhein-neckar.de> for submitting this.
829 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
830 causing dump to block when dumping really huge filesystems.
831 Many thanks to Patrik Schindler <poc@pocnet.net> for
832 helping me find this bug.
834 4. Fixed the treatment for an interrupt signal when dump access
835 the remote tape through RSH. Thanks to Christian Weisgerber
836 <naddy@mips.rhein-neckar.de> for providing the patch.
838 5. Fixed a bug which was causing dump/restore to display
839 garbage characters instead of the remote host name.
841 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
842 =====================================================================
844 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
845 <kukuk@suse.de> for submitting the patch.
847 2. Fix for the exit code when using the size estimate option of
848 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
849 submitting the patch.
851 3. Handle EINTR in atomical reads/writes in dump, which was causing
852 dump to fail on some systems. Thanks to Eric Jergensen
853 <eric@dvns.com> for reporting the bug and submitting the patch.
855 4. Handle more than 16 characters for the device names in dumpdates.
856 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
857 tracking down the problem and proposing the solution.
859 5. Fixed a bug in dump which prevented the creation of the
860 dumpdates file when doing a 0-level dump without already
861 having a dumpdates file. Thanks to Patrik Schindler
862 <poc@pocnet.net> for reporting the bug.
864 6. Changed the way dump 'S' flag reports the size estimate
865 from number of blocks into bytes (making it compatible
866 with the Solaris version, and simplifying things for
867 amanda users). Thanks to Jason L Tibbitts III
868 <tibbs@math.uh.edu> for reporting the bug.
870 7. Fixed a compatibility problem in linux/alpha dump tape format.
871 Now the linux/alpha dump are (again) compatible with the
872 other dump formats. But this breaks compatibility with
873 older dumps made on alpha. Thanks to Mike Tibor
874 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
876 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
877 ======================================================================
879 1. Added a '--enable-kerberos' to configure.
881 2. Added a 'S' option to dump which determines the amount of space
882 that is needed to perform the dump without actually doing it, similar
883 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
884 <cermak@ahab.rutgers.edu>.
886 3. Added a 'M' multi-volume option to dump and restore which enables
887 dumping to multiple files (useful when dumping to an ext2
888 partition to make several dump files in order to bypass the 2GB
889 file size limitation). The 'f' argument is treated as a prefix and
890 the output files will be named <prefix>001, <prefix>002 etc. With
891 the 'M' flag, restore automatically selects the right file without
892 asking to enter a new tape each time.
894 4. Fixed a memory leak which was causing dump to grow very big
895 (270MB when dumping a 10GB filesystem...). Thanks to Jason
896 Fearon <jasonf@netrider.org.au> for reporting the bug.
898 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
899 ======================================================================
901 1. Make configure test if the system glob routines support
902 extended syntax (ALTDIRFUNC). If not, use the internal glob
903 routines instead of system ones. Thanks to Bernhard Sadlowski
904 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
905 and helping me resolve this and other minor libc5 compiling
908 2. Fix a problem when dumping a ext2fs with the 'filetype'
909 feature enabled. Thanks to Patrick J. LoPresti
910 <patl@cag.lcs.mit.edu> for reporting the bug and to
911 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
913 3. Made the nodump flag work on directories. A directory which
914 has the nodump flag gets never dumped, regardless of its
917 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
918 which allows dump on an active ext3 filesystem. However, this
919 is a "quick and dirty" patch which enables backup of an ext3
920 filesystem through the ext2 compatibility (by ignoring the
921 NEEDS_RECOVERY bit). The journal file is not recognized and
922 it is dumped (it should not).
924 5. Test the superblock compatibility flags when dumping, in order
925 to be sure that we know how to deal with specific features.
927 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
928 ====================================================================
930 1. Use lchown instead of chown, fixing a possible security problem
931 when restoring symlinks (a malicious user could use this
932 to deliberately corrupt the ownership of important system files).
933 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
934 this and providing the patch.
936 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
937 ====================================================================
939 1. Put dump sources under CVS, added Id tags in all files so
940 one can use 'ident' on binary files.
942 2. Added the dump/restore version in the usage text so one can
943 easily verify the version he is using.
945 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
946 a va_start/va_end problem on linux-ppc (always call va_start
947 va_end in pairs each time we use a vararg function).
949 4. Added again the DT_* constants because old libc does not
950 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
951 for submitting the bug report.
953 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
954 this should enable dumping big (huge) filesystems.
956 6. Added the RSH environment variable in order to be able to
957 use a rsh replacement like ssh when doing remote backups (and
958 bypass the security limitations of rcmd). Now you can do remote
959 backups without being root (or making dump setuid root).
961 7. Modified again the way dumpdates works. For incremental dumps,
962 we need to read dumpdates even if we are not using 'u' option.
963 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
966 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
967 ===================================================================
969 1. Removed the 'k' flag from the restore 'about' text if kerberos
972 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
973 to them (fsetflags takes a char*, setflags an open fd!).
975 3. (f)setflags is called only if the flags aren't empty. If the
976 file is a special file, a warning is printed, because changing
977 flags implies opening the device. Normally, a special file
978 should not have any flag... (Debian bug #29775, patch provided
979 by Abhijit Dasgupta <abhijit@ans.net>).
981 4. Made possible to dump a file system not mentioned in /etc/fstab.
982 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
984 5. Changed the default behaviour to not create dumpdates
985 unless 'u' option is specified. Removed the old "debian-patch"
986 which provided the same thing. (Debian bug #38136, #33818).
988 6. Removed all those dump*announce, since they were getting old...
990 7. Added warning messages if dumpdates does not exist and
991 when an inferior level dump does not exist (except for a level 0
994 8. Debugged the glob calls in interactive mode: restore used a
995 dirent struct which was different from the /usr/include/dirent.h
996 one (this used to work, is it a glibc2 change?), so none of the
997 compat glob (which used /usr/include/dirent.h) or the system glob
998 worked. Restore use now the system dirent (and the system
999 DT_* constants), which are compatible with BSD ones.
1001 9. Added a configure flag (--with-dumpdatespath) to specify
1002 the location of dumpdates file. By default, it is
1005 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1006 included the current date/version in man pages.
1008 11. Corrected the estimation of remaining time when
1009 the operator doesn't change the tapes quickly enough. This
1010 was an old bug, I thought I corrected it, and discovered
1011 that in fact it was corrected in two different places, so
1012 the results canceled each other...
1014 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1015 ===================================================================
1017 1. Integrated multiple patches from RedHat, Debian and SuSE:
1019 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1021 - several fixes in the man pages.
1022 - update the default tape device to /dev/st0.
1023 - many updates for Linux Alpha (byte ordering, size_t etc).
1025 - use environment variable for TMPDIR (instead of /tmp).
1026 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1027 - workaround egcs bug (RedHat bugs #4281 and #2989).
1028 - wire $(OPT) throughout Makefile's.
1030 2. Upgrade the dump revision to 1, making possible to dump filesystems
1031 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1033 3. Fix some compile warnings, prototype all functions.
1035 4. Use glibc err/glob instead of internal compatibility
1036 routines (only if available).
1038 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1039 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1041 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1042 ======================================================================
1044 1. Integrated the changes from FreeBSD-3.1-RELEASE
1045 (mostly bug fixes, buffer overruns, dump has now an "automatic
1046 tape length calculation" flag, dump/restore can use kerberos now
1047 (this is NOT tested), use environment variables for TAPE and
1050 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1052 3. Corrected a bug in restore when using 'C' option with multi-volumes
1053 tapes (files splited accros two tapes give "size changed" errors
1056 4. Corrected the long standing bug when dumping multiple tapes.
1057 This works for me, needs further testing.
1059 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1060 ====================================================================
1062 1. Dump now runs correctly on kernels 2.1.x
1063 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1065 Changes between versions 0.4b2 and 0.4b3
1066 ========================================
1068 1. Use realpath() if available
1070 2. Report statistics
1072 Changes between versions 0.4b1 and 0.4b2
1073 ========================================
1075 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1078 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1080 3. Fixed configure.in to correctly handle --enable-rmt
1082 Changes between versions 0.3 and 0.4b1
1083 ======================================
1085 1. Integrated the changes from 4.4BSD-Lite2
1087 2. Integrated the patches from Debian and Red Hat
1089 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1091 4. Changed dump to use the Ext2fs library to get block addresses. This
1092 should solve the endianness problem on SparcLinux.
1094 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1095 distribution's one) to use autoconf
1097 6. Fixed a few minor bugs
1099 Changes between versions 0.2e and 0.2f
1100 ======================================
1102 1. Added the creation of named pipes (fifos) in restore.
1104 2. Added the -N flag in the restore manual page.
1106 3. Added the file kernel.patch which contains the llseek() optimization
1107 patch for 1.2.x kernels.
1109 4. Fixed a bug in the restoration of symbolic links: owner and group were
1112 5. Integrated some changes from FreeBSD 2.2.
1114 6. Added a call to ftruncate() after restoring each file to restore
1115 correctly files ending by a hole.
1117 Changes between versions 0.2d and 0.2e
1118 ======================================
1120 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1121 with owner = root (0) and group = root/wheel/whatever (0).
1123 Changes between versions 0.2c and 0.2d
1124 ======================================
1126 1. Dump is now able to backup 2GB+ filesystems.
1128 2. Dump and restore can now be linked as static binaries.
1130 Changes between versions 0.2b and 0.2c
1131 ======================================
1133 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1135 Changes between versions 0.2a and 0.2b
1136 ======================================
1138 1. Really fixed the bug that I should have corrected in 0.2a.
1140 2. Enabled optimization again.
1142 Changes between versions 0.2 and 0.2a
1143 =====================================
1145 1. Disabled the optimization during compilation.
1147 Changes between versions 0.1 and 0.2
1148 ====================================
1150 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1151 the fs_type field (actually, I modified the file fstab.c to make it
1152 use the mntent functions).
1154 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1155 made by dump should be readable by the BSD restore and backups made
1156 by the BSD dump should be readable by restore. Unfortunately, this
1157 means that the dump format has changed between version 0.1 and version
1160 3. Dump is now able to backup a subtree, it is no longer limited to whole
1161 filesystems like the BSD version.
1163 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1166 Changes between versions 0.0 and 0.1
1167 ====================================
1169 1. Now create links rdump and rrestore during the `make install' step.
1171 2. Linux port specific bugs added to the manual pages
1173 3. Incorrect estimation of the number of tapes blocks fixed when doing
1174 incremental backups.
1176 4. Better ls-like format in restore in interactive mode.