1 $Id: CHANGES,v 1.203 2002/12/12 11:49:35 stelian Exp $
3 Changes between versions 0.4b32 and 0.4b33 (released ?????????????????)
4 =======================================================================
6 1. Added a note in the restore man page clarifying the question
7 'set the permissions on the current directory ?' asked by
8 restore at the end of treatment in -i and -x modes.
10 2. Fixed the endianess issues when reading compressed tapes.
11 Thanks to Dark Force <daq4th@users.sourceforge.net> for
12 reporting this bug and providing test cases.
14 3. Fixed the "ACL won't be dumped" warning message (which showed
15 an extra, unrelated error message). Thanks to Dragan Krnic
16 <dkrnic@lycos.com> for reporting this bug.
18 4. Made dump look first into /etc/fstab, then into /etc/mtab
19 when searching for filesystem paths. Also fixed some problems
20 caused by binding mounts or multiple block device mounts.
21 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
22 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>.
23 (fixes Debian bugs #147086 and #160305, Sourceforge bug #618699)
25 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
26 =======================================================================
28 1. Changed dump to use fcntl(F_SETLK) style locking instead
29 of flock() when locking the dumpdates file. With the old
30 locking scheme, a local user having read rights on the
31 dumpdates file could be able to do a Denial of Service attack
32 on dump. In order to lock the dumpdates file with the new
33 scheme, the user would need to have write access on the file.
34 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
35 reporting the bug (originally a bugtraq post).
37 2. Fixed interactive 'ls' which caused spurious errors warnings
38 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
39 <jorgen_ostling@users.sourceforge.net> for reporting this
42 3. Fixed dump's estimate when dealing with sparse inodes.
44 4. Modified dump to allow setting a blocksize bigger than 32kB
45 (raised the limit to 1024kB), because newer hardware needs
46 this for proper operation. Thanks to Dirk Traenapp
47 <dtraenapp@users.sourceforge.net> for reporting this.
49 5. Fixed a bug causing Dump to stop and report an error if an
50 inode number in the exclude file was followed by some amount
51 of whitespace. Thanks to Jeffrey Sofferin
52 <sofferin@users.sourceforge.net> for reporting this bug.
54 6. Fixed a bug which caused restore, in some particular cases,
55 to ask some 'scary' questions and leave a bunch of RSTTMP
56 directories behind when restoring incremental tapes. Thanks
57 to Philippe Troin <phil@fifi.org> for reporting this bug and
58 providing the test cases.
60 7. Changed the wording when inodes are excluded from dump:
61 replaced 'Added inode 7 to exclude list' with
62 'Excluding inode 7 (resize inode) from dump', as suggested
63 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
65 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
66 ===================================================================
68 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
69 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
70 <eros@lamel.bo.cnr.it> for reporting the bug.
72 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
73 ===================================================================
75 1. Made dump print out the ext2 logical block number in case of a read
76 error, which can be given as an argument to debugfs' ncheck command,
77 in order to find out the affected inode. Added note about this
78 usage in dump's man page.
80 2. Fixed a problem in restore when reading tapes written on big
81 endian machines with very old versions of dump. The patch was
82 contributed by George Helffrich <george@geology.bristol.ac.uk>.
84 3. Fixed the tape length calculation when using large tapes
85 and compression. Thanks to Georg Lippold
86 <g_lippold@sourceforge.net> for reporting the bug.
88 4. Added a new set of examples in dump_on_cd_2 directory, based
89 on dump_on_cd examples but somewhat enhanced, supporting DVD
90 media, and localized in english and german. Thanks to
91 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
93 5. Made dump save 32 bit UID/GID. Internally, this was achieved
94 by switching from the old BSD inode format to the new BSD
95 inode format, which means that the tape format was changed.
96 However, since all restore versions out there should
97 transparently support both inode formats, the change should
98 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
101 6. Fixed a lot of warnings in the code shown when compiling
102 with 'gcc -W'. Thanks to Matthias Andree
103 <matthias.andree@stud.uni-dortmund.de> for reporting this.
105 7. Fixed a small markup bug in the dump man page. Thanks to
106 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
109 8. Rewrote entirely the man pages using the tmac.an macro
110 package (Linux man page format) instead of the original BSD
111 format. They should be now cleaner and easier to modify.
113 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
114 ==================================================================
116 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
117 non Linux clients were misinterpreted. The description of the problem
118 (incompatible numbering in Linux mtio opcodes) is documented at
119 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
120 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
121 providing an excellent, cross-platform replacement for rmt in his
124 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
125 which caused restore to fail when the inode maps were not entirely
126 contained on the first volume (when using really small volumes or
127 when dumping a huge number of inodes).
129 3. Fixed a problem in dump, where files in subdirectories of directories
130 that have the nodump flag set (or being excluded with -e/-E) were
131 sometimes incorrectly dumped (instead of being excluded). The original
132 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
133 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
134 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
135 help testing the patch.
137 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
138 which makes one able to pipe the output of dump, by the net, to
139 a remote CD-burner server.
141 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
142 and use GNU's symbolic syntax over rmt instead of numerical values
143 to assure multiple platform compatibility.
145 6. Documented the -d option in restore.
147 7. Added a -v (verbose) mode to dump. For now it just prints the number
148 of the inode being dumped, but this could evolve in future versions
149 to include interesting debugging output.
151 8. Added a -o flag to restore, which automatically restores the current
152 directory permissions in -i or -x mode, without asking the operator
153 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
154 and Peter Samuel <peters@e-smith.com>.
156 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
157 ====================================================================
159 1. Fixed a bug in the QFA feature which could have caused
160 a bad tape positionning on restore (causing extra delays in
161 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
163 2. Added a small note in the dump man page specifying that
164 there should be no white space between the option letter and
165 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
167 3. Made restore work with older versions of the readline library,
168 by checking in configure for several extended symbols. Restore
169 can now be compiled with a readline as old as the 2.0 release
170 (though it may be a good idea to upgrade it to a more recent
171 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
172 reporting the build failures.
174 4. Fixed a performance problem with the QFA file creation in
175 dump, which made unnecessary seeks on the tape slowing down
176 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
177 reporting this issue.
179 5. Removed the inclusion of some kernel headers in the dump
180 source, which prevented the compile in some kernel/glibc
181 headers/architecture combination. Thanks to Bdale Garbee
182 <bdale@gag.com> for reporting the bug.
184 6. Added the appropriate error message when dump fails to
185 open the output file for writing. Thanks to Amith Varghese
186 <amithv@yahoo.com> for reporting this bug.
188 7. Made restore able to understand large Solaris ufsdump tapes
189 (containing inodes bigger than 4194304). Sun have introduced
190 an "extension" to the dump tape format when dealing with
191 those inodes, which was uncorrectly handled by Linux restore.
192 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
193 providing a test case.
195 8. Added the -m parameter to dump which optimises the output for
196 inodes having been changed but not modified since the last dump
197 ('changed' and 'modified' have the meaning defined in stat(2)).
198 For those inodes, dump will save only the metadata, instead of
199 saving the entire inode contents. Inodes which are either
200 directories or have been modified since the last dump are saved
201 in a regular way. Uses of this flag must be consistent, meaning
202 that either every dump in an incremental dump set have the flag,
205 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
206 =======================================================================
208 1. Fixed behaviour of dump when exceeding resource limits
211 2. Added the -L flag to restore to allow the user to specify a
212 maximal allowed number of miscompares when using restore
213 with the -C option to check the backup.
215 3. Detailed the manual entry for the -N option of restore.
217 4. Added the -a flag to restore to make able doing unattended
218 restores in -i or -x mode (automatically walks through the
219 multiple dump volumes).
221 5. Extended the QFA mode to work with local files and/or
222 remote tapes and files. This way, restore can know in advance
223 the tape number and the offset for the inodes to extract and
224 can minimize the extraction time by seeking directly to the
227 6. Added the -A <archive> option to both dump and restore,
228 which makes dump to archive a dump table-of-contents in
229 the specified file to be used by restore to determine
230 whether a file is in the dump file that is being restored.
231 (the archive file syntax is also compatible with the
232 Solaris ufsdump generated one).
234 7. Small fix in restore making it able to read some (broken ?)
235 Solaris ufsdump tapes.
237 8. Fixed dump to correctly recognise the root filesystem when using
238 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
239 for reporting this bug.
241 9. Added the -P <file> option to restore to create a
242 Quick File Access file from an already made dump. Patch
243 contributed by Uwe Gohlke <uwe@ugsoft.de>.
245 10. Made restore compile and run on Solaris, making it a
246 possible replacement for the standard ufsrestore. Port was
247 contributed by Uwe Gohlke <uwe@ugsoft.de>.
249 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
250 =====================================================================
252 1. Added a set of backup scripts from Eugenio Diaz
253 <getnito@yahoo.com> in the examples section. It features
254 automatic (cron based) full and incremental dumping of
255 several filesystems on a separate filesystem.
257 2. Fixed a off-by-one miscalculation which disabled dumping
258 a one letter subdirectory. Thanks to Chris Siebenmann
259 <cks@utcc.utoronto.ca> for reporting the bug.
261 3. Fixed several restore bugs occuring when trying to
262 restore some missing files on the tape. Thanks to Chris
263 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
265 4. Fixed --with-ldopts configure argument passing, installing from
266 a separate object directory, makefile cleanups contributed
267 by <splite@purdue.edu>.
269 5. Fix a bug which could caused, in some conditions, the highest
270 number inode of a filesystem, to not be dumped. Many thanks
271 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
274 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
275 =======================================================================
277 1. Added a mini howto from Patrick Walsh in the examples
280 2. Minor man pages syntax corrections. Thanks to
281 Chris Danis <screechco@home.com> for reporting the bugs.
283 3. Added a script from David B. Peterson <dave@toppledwagon.com>
284 to the examples section. It features dumping several
285 filesystems to a remote tape drive upon ssh.
287 4. Added a patch provided by Richard Jones <rich@annexia.org>
288 which allows BRADEMAX (number of read errors tolerated by
289 dump) to be adjusted using the -I option.
291 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
292 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
295 6. Add the -l option to restore to specify if, when doing a
296 remote restore, the file used to access the backup is a
297 regular file (the defaults being a tape drive). Restore needs
298 to know this information when reading a remote compressed
299 dump. Previously, this information was autodetected, but
300 the autodetection code fails (with ioctl: Inappropriate ioctl
301 for device) when using a non Linux remote box. Thanks to
302 many users and especially to Eros Albertazzi
303 <eros@lamel.bo.cnr.it> for reporting this.
305 7. Found a workaroung for the dump deadlock problem (3 childs
306 stuck in pause(), father in read()). The workaround seems
307 to work for me and several beta-testers. If it doesn't work
308 for you, please report back.
310 8. Updated the RPM spec file (BuildPrereq, URL etc).
312 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
313 ========================================================================
315 1. Fixed the permissions of a newly created QFA file by dump.
317 2. Cleaned up the source of dump (the external variables
318 definition was a complete mess, making possible to have
321 3. Fixed restore to use the full tape volume path when doing
322 a compare (since it changes the working directory to the
323 filesystem being compared in the process).
325 4. Added the -q option to dump which makes dump abort
326 whenever it needs operator attention. It should be
327 easier to use dump in scripts with this option.
329 5. Detect the use of incompatible options to dump and
330 refuse them (like -a and -B options together).
332 6. Added bzip2 compression to dump/restore (use option -j level
333 to select it). Note that you will need a restore version
334 >= 0.4b24 in order to restore a bzip2 compressed dump.
335 The same warning as for the zlib compression applies:
336 the tape format of a bzip2 dump is not compatible with the
337 original BSD tape format.
339 7. Fixed a overflow problem in dump corrupting the dump when
340 very large files were encountered. Thanks to Vladimir Nadvornik
343 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
344 kernel buffer/page cache before starting the dump, helping
345 a bit those who use dump on mounted filesystems. Thanks to
346 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
349 9. Updated the RPM spec file following the RedHat changes
350 (dynamically linked binaries now in /usr/sbin etc).
352 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
353 which enables restore to recognise multi volume compressed dumps
354 done on CD. Included his scripts for dump (which pipe the dump
355 output directly into cdrecord) and restore. There is now
356 possible to to backups to CD on the fly!
358 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
359 ===================================================================
361 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
362 Marcus Meissner (Caldera International Security Dept.).
364 2. Implement the Sun rmt extended protocol. Patch submitted
365 by Ian Gordon <iangordon@users.sourceforge.net>.
367 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
368 the linux kernel header. This ensures that dump always has the
369 latest version of this file. Patch submitted by
370 Andreas Dilger <adilger@turbolinux.com>.
372 4. Report any filesystem present in either /etc/fstab with a
373 positive passno or /etc/dumpdates in dump -w output.
374 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
376 5. Fixed the looping problem in dump introduced in the
379 6. Changed the -B option of dump to limit the size of
380 _compressed_ output per volume if compression is on.
381 Patch contributed by Helmut Jarausch
382 <jarausch@igpm.rwth-aachen.de>. Note however that, since
383 it is impossible to predict the size of the compressed
384 data before doing the compression, the -B limit is a bit
387 7. Fixed a bug in reading the operator typed file/tape path for
388 the next volume in restore.
390 8. Implemented a "-F script" option to restore which permits the
391 user to specify a script which will be launched at the
392 beginning of each tape, useful for automatic programming of
393 tape changers for example. See the restore man page for the
394 script parameters and return codes.
396 9. Small fix for the QFA routines provided by Uwe Gohlke
397 <uwe@ugsoft.de>, and some recommendations for QFA uses in
400 10. Fixed the multivolume restoring where making a mistake
401 to the 'Mount next tape' prompt caused several blocks to
404 11. Enhanced the -e option of dump to take as a parameter a
405 comma separated list of inode numbers.
407 12. Added the -E option to dump which specify a file containing
408 inode numbers to exclude from the dump.
410 13. Fixed the compressed multi-volume dump + restore.
412 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
413 ==================================================================
415 1. Made dump report the number of blocks dumped per volume.
416 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
418 2. Fix a bug in dump where waiting too much at the 'change volumes'
419 question caused the volume to be rewritten. Thanks to
420 Graham Reed <greed@users.sourceforge.net> for reporting the
421 bug and providing a patch.
423 3. Added a compression option to dump, contributed by Jerry
424 Peters <gapeters@worldnet.att.net>.
426 WARNING: the compression works only on files/pipes and
427 drives supporting variable block size.
429 WARNING2: tapes created using the compression feature are
430 incompatible with the standard dump tape format, and a
431 version of restore >= 0.4b22 will be needed for extraction.
433 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
434 architectures. Thanks to Paul Slootman <paul@debian.org> for
435 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
438 5. Many cleanups (CPP defines, const char warnings, check of
439 ext2fs COMPAT flags, time_t cleanups, added libext2 version
440 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
442 6. Made --prefix option work in configure. All the install path
443 are now based on the configure parameters.
445 7. Added the Quick File Access mode in dump/restore, contributed
446 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
447 a file tape position for each inode, and this file is used by
448 restore (if called with parameter Q and the filename)
449 to directly position the tape at the file restore is currently
450 working on. This saves hours when restoring single files from
451 large backups, saves the tapes and the drive's head. Use
452 --enable-qfa option of configure to compile in the QFA support.
454 8. Added the possibility to dump several files and directories
455 in a single invocation of dump. Thanks to Uwe Gohlke
456 <uwe@ugsoft.de> for implementing this option.
458 9. Fixed the dumping and restoring of files with holes
459 (files ending with holes didn't get dumped or restored
462 10. Fixed a socket descriptor leak in dump, which leaved opened
463 3 file descriptors per dump process (and there is one dump
466 11. Fixed dump large file system support, by explicit use of
467 open64/lseek64/etc functions (explicit use needed because
468 e2fsprogs libraries don't behave well when compiled with
469 FILE_OFFSET_BITS=64).
471 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
472 ======================================================================
474 1. Fixed some bugs in the dump -w|-W logic introduced by
475 the previous version. Thanks to Andreas Dilger
476 <adilger@turbolinux.com> for his help on this one.
478 2. Fixed again a compilation problem when using old e2fs
479 headers (filesystem label related). Thanks to many users
480 who reported this stupid error.
482 3. Fixed a build problem on old lib5 based systems dealing with
483 _PATH_MNTTAB being undefined. Thanks to John Adams
484 <johna@onevista.com> for reporting the bug.
486 4. Improved the error detection in restore when creating
487 the temporary files in TMPDIR. Now restore will corectly
488 report a 'no space left on device' error instead of
489 strange errors which could imply an archive corruption.
490 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
491 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
493 5. Added the throughput information in dump's progression
494 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
497 6. Use libext2fs's inode scanning routines, which are particularly
498 robust when dealing with errors (especially when having some
499 bad blocks on the filesystem). Many thanks to Colin
500 <colin@horizon.com> for reporting the bug, investigating
501 the issues, finding a workaround, writing the patch and
502 fully testing it... (of course, if this patch will break
503 anything, he is to blame for :-)).
505 7. Made dump and restore LFS aware. Dump can dump a filesystem
506 containing large files, generate a large file on output and
507 restore can restore them. This feature is not enabled by
508 default, use --enable-largefile option of configure to enable
509 it (you need to have a LFS aware glibc though). Thanks to
510 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
511 and to Theodore T'so <tytso@valinux.com> for his always
514 8. Made dump ask upon a tape write error if it should rewrite
515 the current volume (assume this is a bad tape) or if it should
516 assume an end-of-tape condition (useful for tape drives which
517 behaves badly at the end of the tape). Thanks to Andreas
518 Dilger <adilger@turbolinux.com> for the suggestion.
520 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
521 =======================================================================
523 1. Fixed a small compilation problem due to a change
524 in the definintion of the struct sigaction in
525 glibc 2.0 / libc5. Thanks to Gunther Schlegel
526 <schlegel@riege.de> for reporting the bug and to
527 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
530 2. Modified the label and uuid analysis in order to be
531 self-contained (independant of kernel/libc headers). This
532 should fix the compile with older kernel/libc headers and
533 will preserve the functionality. Thanks to Bernhard Erdmann
534 <bernhard.erdmann@gmx.de> for reporting the bug.
536 3. The 'exclude inode' option, if applied to a directory
537 inode, excludes now correctly all the leaves of this
538 directory. Thanks to John R. Dennison
539 <gerdesas@users.sourceforge.net> for reporting the bug.
541 4. Fixed the '-e' option to disable the possibility
542 to exclude the root inode (which causes the dumps to
543 be unreadable by restore). Prevented array overflow
544 when multiple -e options are used.
546 5. Fixed dump to correctly interpret a filesystem argument
547 which is a mountpoint and it is not an absolute path
548 (as specified in the fstab). Thanks to Bernhard R. Erdmann
549 <be@berdmann.de> for reporting the bug.
551 6. Made dump able to backup files larger than 2 GB. Note that
552 dump still doesn't cope with files larger than 4 GB.
554 7. Restore the real uid and gid before invoking an external
555 RSH program (big hole when dump or restore is suid root!).
557 8. Get the values for the default tape device and the location
558 of fstab file from the system headers. Thanks to
559 Andreas Dilger <adilger@turbolinux.com> for the patch.
561 9. Made dump -w|-W report all recognized filesystems
562 present in either /etc/fstab or /etc/dumpdates, and present
563 the list in the same order as in fstab file. Thanks
564 to Andreas Dilger <adilger@turbolinux.com> for the patch.
566 10. Made dump's -a (automatic end of tape detection) the
567 default. Specifying one of -B, -c, -d or -s options will
568 override the EOM detection. Thanks to Andreas Dilger
569 <adilger@turbolinux.com> for the patch.
571 11. Save the ext2 filesystem label into the dump volume label.
572 Specifying a volume label on the command line (-L option)
573 will override this feature. Thanks to Andreas Dilger
574 <adilger@turbolinux.com> for the patch.
576 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
577 =====================================================================
579 1. Fixed the signal handling in dump (which I broke in 0.4b17)
580 which was causing several strange dump failures (dump
581 hanged or segmentation faults etc).
583 2. Specified the default blocksize in the dump man page.
585 3. Changed two info messages of restore to be written on stdout
586 instead of stderr in order to leave stderr only for errors
587 or warnings. Thanks to Stephen Atwell
588 <satwell@urbana.css.mot.com> for the suggestion.
590 4. Corrected an off by one calculation which prevented
591 dumping a directory having a 1 character name length.
592 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
593 for reporting the bug.
595 5. Reinforce dump in order to allow only level 0 and no
596 -u option when dumping a subdirectory, and document
597 this in the man page. Thanks to Bernhard Erdmann
598 <bernhard.erdmann@gmx.de> for reporting the bug.
600 6. Fixed a small harmless bug in restore which caused
601 in some conditions hard links to be restored several
602 times (therefore generation some warning messages).
603 Thanks to Kenneth Porter <shiva@well.com> for
606 7. Updated the RPM spec file to the latest RedHat version,
607 providing FHS packaging and other cosmetic changes.
608 You will need RPM version >= 3.0.5 in order to build the RPM.
610 8. Updated the configure script to check for libtermcap
611 before checking for libreadline, because we need this
612 library in order to compile the readline support.
614 9. Made dump understand the LABEL= and UUID= notation
615 both in /etc/fstab and on the command line. Note that
616 you will need the /proc filesystem in order to use
617 these notations. Thanks to Erik Troan <ewt@redhat.com>
618 for providing the patch.
620 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
621 ===================================================================
623 1. Fixed a potential buffer overflow in restore. Thanks
624 to Stan Bubrouski <satan@fastdial.net> for reporting
627 2. Fixed a readline-related bug which prevented
628 'cat DUMPFILE | restore -i -f -' from working. Thanks
629 to Charles Karney <karney@users.sourceforge.net>
632 3. Changed a few "panic" into "exit", causing restore to
633 be more stable against some attacks (like the last one
634 reported on Bugtraq, although the last version of restore
635 was not vulnerable - just dumped core). Thanks to
636 Andreas Hasenack <andreas@conectiva.com.br> for reporting
639 4. Removed the suid-root bit on dump and restore in the
640 default build (and generated RPMs). It should be safer
641 now. Users who need the suid-root capabilities in order
642 to make network backups should read first the man page
643 of dump and enable the suid bit manually.
645 5. Added -ltermcap to the compile parameters for restore
646 when using readline support, in order to make the compile
647 process work on some systems (SuSE ?). Thanks to
648 Patrik Schindler <poc@pocnet.net> for reporting the bug.
650 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
651 ====================================================================
653 1. The -F script is called now *only* at the end of a tape,
654 not at the end of the dump. Thanks to Christian Haul
655 <haul@informatik.tu-darmstadt.de> for the bug report.
657 Normally, the device name and the current volume number
658 are passed on the command line of the script. However,
659 if you want the old style script (with no arguments
660 passed on the command line), you can enable it in
661 configure with the --enable-oldstylefscript.
663 2. Use posix signal handling to preserve dump functionality
664 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
667 3. Made the exit code of restore in case of a 'C'ompare
668 command reflect the result of the compare. An exit status
669 of 0 means the dump archive is correct, 1 means tape errors,
670 2 means that some files were modified. Thanks to Kenneth Porter
671 <shiva@well.com> for the suggestion.
673 4. Made (finally) quotes work in interactive restore.
675 5. Small fixes in order to allow dump to compile with a
676 really old e2fsprogs version. Thanks to Ian Zimmerman
677 <itz@speakeasy.org> for the bug report.
679 6. Add GNU readline capabilities to interactive restore.
680 Use configure's --enable-readline flag to enable this feature.
681 Thanks to Patrik Schindler <poc@pocnet.net> for the
684 7. Do the compare on the fly in restore 'C' mode (this will
685 allow not to exhaust the available /tmp space when
686 ccmparing large files). Thanks to Kenneth Porter
687 <shiva@well.com> for the suggestion.
689 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
690 ====================================================================
692 1. Fixed some several duplicate 'DUMP: DUMP:' in the
695 2. Corrected the estimation of blocks to dump. Note that
696 this is still wrong for large directory inodes, due
697 to the size of a BSD directory entry on the tape
698 (estimation is based on the size of the ext2 directory,
699 which is smaller because it doesn't contain the null
700 character at the end).
702 3. Made dump report the total number of MB written to
703 tape at the end of run. Thanks to W. Reilly Cooley
704 <wcooley@nakedape.cc> for the patch.
706 4. Added the -X option to restore giving the possibility
707 to read the names of the files to be extracted/listed
708 from a text file (in addition of the command line).
709 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
712 5. Added the device name and the current volume number
713 as arguments to the end of tape script (dump -F option).
715 6. Made the multi-volume dump work again (after having
716 broken it in 0.4b15).
718 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
719 ===================================================================
721 1. Added a prompt command in interactive restore mode. Thanks
722 to Andreas Dilger <adilger@home.com> for the patch.
724 2. Fixed a buffer overflow problem in dump (caused by
725 not checking the size of the filesystem parameter).
726 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
727 reporting this on Bugtraq (and to several dump users
728 who forwarded me his mail).
730 3. Added the '-F script' option to dump in order to
731 launch a script at the end of each tape (to be used
732 with a tape changer, or to notify the sysadmin by
735 4. Fixed a bug in restore compare code caused by the changes
738 5. Fixed the treatment of options using the old BSD syntax
739 in both dump and restore.
741 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
742 =======================================================================
744 1. Fixed a bug in dump which may have caused invalid deleted
745 directories to be dumped out if they were deleted after the
746 mapping pass. This could occure on active filesystem and lead
747 to heap corruption (causing dump malfunction in many possible ways).
748 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
749 bug and submitting the patch.
751 2. Fixed the handling of the filesystem parameter in dump. A
752 '/mnt/disk' parameter caused the disk contents to be dumped,
753 but a '/mnt/disk/' parameter caused the mountpoint directory
754 to be dumped (generally an empty directory).
756 3. Improved the output of dump in order to tell which directory
757 it is currently dumping (when dumping a subtree).
759 4. Added the '-e' exclude inode option to dump. Thanks to
760 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
762 5. Added a REPORTING-BUGS file in order to provide a guide
763 on how to correctly report a bug in dump/restore.
765 6. Fixed a restore problem when restoring a hard link to an inode
766 having the immutable or append-only attribute set. Thanks to
767 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
769 7. Fixed a compatibility problem between dumps made on little
770 endian machines (the format was correct) and big endian
771 machines (the format was incorrect). This fix break the
772 compatibility with the older dumps made on big endian
773 machines (sparc, m86k, ppc etc). For the first time in
774 linux dump's history (I believe), the dumps made by *BSD,
775 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
776 so interchangeable. Thanks to Rob Cermak
777 <cermak@ahab.rutgers.edu> for submitting the bug and
778 helping me test the fix.
780 8. Fixed the way dump reports the remaining percent/time, if
781 the number of blocks actually dumped exceeds the estimated
782 number of blocks. Thanks to Jean-Paul van der Jagt
783 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
785 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
786 ======================================================================
788 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
789 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
791 2. Removed the requirement to build the RPM as root from the
792 spec file. Thanks to Christian Weisgerber
793 <naddy@mips.rhein-neckar.de> for submitting this.
795 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
796 causing dump to block when dumping really huge filesystems.
797 Many thanks to Patrik Schindler <poc@pocnet.net> for
798 helping me find this bug.
800 4. Fixed the treatment for an interrupt signal when dump access
801 the remote tape through RSH. Thanks to Christian Weisgerber
802 <naddy@mips.rhein-neckar.de> for providing the patch.
804 5. Fixed a bug which was causing dump/restore to display
805 garbage characters instead of the remote host name.
807 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
808 =====================================================================
810 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
811 <kukuk@suse.de> for submitting the patch.
813 2. Fix for the exit code when using the size estimate option of
814 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
815 submitting the patch.
817 3. Handle EINTR in atomical reads/writes in dump, which was causing
818 dump to fail on some systems. Thanks to Eric Jergensen
819 <eric@dvns.com> for reporting the bug and submitting the patch.
821 4. Handle more than 16 characters for the device names in dumpdates.
822 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
823 tracking down the problem and proposing the solution.
825 5. Fixed a bug in dump which prevented the creation of the
826 dumpdates file when doing a 0-level dump without already
827 having a dumpdates file. Thanks to Patrik Schindler
828 <poc@pocnet.net> for reporting the bug.
830 6. Changed the way dump 'S' flag reports the size estimate
831 from number of blocks into bytes (making it compatible
832 with the Solaris version, and simplifying things for
833 amanda users). Thanks to Jason L Tibbitts III
834 <tibbs@math.uh.edu> for reporting the bug.
836 7. Fixed a compatibility problem in linux/alpha dump tape format.
837 Now the linux/alpha dump are (again) compatible with the
838 other dump formats. But this breaks compatibility with
839 older dumps made on alpha. Thanks to Mike Tibor
840 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
842 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
843 ======================================================================
845 1. Added a '--enable-kerberos' to configure.
847 2. Added a 'S' option to dump which determines the amount of space
848 that is needed to perform the dump without actually doing it, similar
849 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
850 <cermak@ahab.rutgers.edu>.
852 3. Added a 'M' multi-volume option to dump and restore which enables
853 dumping to multiple files (useful when dumping to an ext2
854 partition to make several dump files in order to bypass the 2GB
855 file size limitation). The 'f' argument is treated as a prefix and
856 the output files will be named <prefix>001, <prefix>002 etc. With
857 the 'M' flag, restore automatically selects the right file without
858 asking to enter a new tape each time.
860 4. Fixed a memory leak which was causing dump to grow very big
861 (270MB when dumping a 10GB filesystem...). Thanks to Jason
862 Fearon <jasonf@netrider.org.au> for reporting the bug.
864 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
865 ======================================================================
867 1. Make configure test if the system glob routines support
868 extended syntax (ALTDIRFUNC). If not, use the internal glob
869 routines instead of system ones. Thanks to Bernhard Sadlowski
870 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
871 and helping me resolve this and other minor libc5 compiling
874 2. Fix a problem when dumping a ext2fs with the 'filetype'
875 feature enabled. Thanks to Patrick J. LoPresti
876 <patl@cag.lcs.mit.edu> for reporting the bug and to
877 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
879 3. Made the nodump flag work on directories. A directory which
880 has the nodump flag gets never dumped, regardless of its
883 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
884 which allows dump on an active ext3 filesystem. However, this
885 is a "quick and dirty" patch which enables backup of an ext3
886 filesystem through the ext2 compatibility (by ignoring the
887 NEEDS_RECOVERY bit). The journal file is not recognized and
888 it is dumped (it should not).
890 5. Test the superblock compatibility flags when dumping, in order
891 to be sure that we know how to deal with specific features.
893 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
894 ====================================================================
896 1. Use lchown instead of chown, fixing a possible security problem
897 when restoring symlinks (a malicious user could use this
898 to deliberately corrupt the ownership of important system files).
899 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
900 this and providing the patch.
902 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
903 ====================================================================
905 1. Put dump sources under CVS, added Id tags in all files so
906 one can use 'ident' on binary files.
908 2. Added the dump/restore version in the usage text so one can
909 easily verify the version he is using.
911 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
912 a va_start/va_end problem on linux-ppc (always call va_start
913 va_end in pairs each time we use a vararg function).
915 4. Added again the DT_* constants because old libc does not
916 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
917 for submitting the bug report.
919 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
920 this should enable dumping big (huge) filesystems.
922 6. Added the RSH environment variable in order to be able to
923 use a rsh replacement like ssh when doing remote backups (and
924 bypass the security limitations of rcmd). Now you can do remote
925 backups without being root (or making dump setuid root).
927 7. Modified again the way dumpdates works. For incremental dumps,
928 we need to read dumpdates even if we are not using 'u' option.
929 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
932 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
933 ===================================================================
935 1. Removed the 'k' flag from the restore 'about' text if kerberos
938 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
939 to them (fsetflags takes a char*, setflags an open fd!).
941 3. (f)setflags is called only if the flags aren't empty. If the
942 file is a special file, a warning is printed, because changing
943 flags implies opening the device. Normally, a special file
944 should not have any flag... (Debian bug #29775, patch provided
945 by Abhijit Dasgupta <abhijit@ans.net>).
947 4. Made possible to dump a file system not mentioned in /etc/fstab.
948 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
950 5. Changed the default behaviour to not create dumpdates
951 unless 'u' option is specified. Removed the old "debian-patch"
952 which provided the same thing. (Debian bug #38136, #33818).
954 6. Removed all those dump*announce, since they were getting old...
956 7. Added warning messages if dumpdates does not exist and
957 when an inferior level dump does not exist (except for a level 0
960 8. Debugged the glob calls in interactive mode: restore used a
961 dirent struct which was different from the /usr/include/dirent.h
962 one (this used to work, is it a glibc2 change?), so none of the
963 compat glob (which used /usr/include/dirent.h) or the system glob
964 worked. Restore use now the system dirent (and the system
965 DT_* constants), which are compatible with BSD ones.
967 9. Added a configure flag (--with-dumpdatespath) to specify
968 the location of dumpdates file. By default, it is
971 10. Added the "AUTHOR" and "AVAILABILITY" sections and
972 included the current date/version in man pages.
974 11. Corrected the estimation of remaining time when
975 the operator doesn't change the tapes quickly enough. This
976 was an old bug, I thought I corrected it, and discovered
977 that in fact it was corrected in two different places, so
978 the results canceled each other...
980 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
981 ===================================================================
983 1. Integrated multiple patches from RedHat, Debian and SuSE:
985 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
987 - several fixes in the man pages.
988 - update the default tape device to /dev/st0.
989 - many updates for Linux Alpha (byte ordering, size_t etc).
991 - use environment variable for TMPDIR (instead of /tmp).
992 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
993 - workaround egcs bug (RedHat bugs #4281 and #2989).
994 - wire $(OPT) throughout Makefile's.
996 2. Upgrade the dump revision to 1, making possible to dump filesystems
997 made with e2fsprogs-1.15 or newer. Nothing seems to break...
999 3. Fix some compile warnings, prototype all functions.
1001 4. Use glibc err/glob instead of internal compatibility
1002 routines (only if available).
1004 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1005 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1007 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1008 ======================================================================
1010 1. Integrated the changes from FreeBSD-3.1-RELEASE
1011 (mostly bug fixes, buffer overruns, dump has now an "automatic
1012 tape length calculation" flag, dump/restore can use kerberos now
1013 (this is NOT tested), use environment variables for TAPE and
1016 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1018 3. Corrected a bug in restore when using 'C' option with multi-volumes
1019 tapes (files splited accros two tapes give "size changed" errors
1022 4. Corrected the long standing bug when dumping multiple tapes.
1023 This works for me, needs further testing.
1025 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1026 ====================================================================
1028 1. Dump now runs correctly on kernels 2.1.x
1029 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1031 Changes between versions 0.4b2 and 0.4b3
1032 ========================================
1034 1. Use realpath() if available
1036 2. Report statistics
1038 Changes between versions 0.4b1 and 0.4b2
1039 ========================================
1041 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1044 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1046 3. Fixed configure.in to correctly handle --enable-rmt
1048 Changes between versions 0.3 and 0.4b1
1049 ======================================
1051 1. Integrated the changes from 4.4BSD-Lite2
1053 2. Integrated the patches from Debian and Red Hat
1055 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1057 4. Changed dump to use the Ext2fs library to get block addresses. This
1058 should solve the endianness problem on SparcLinux.
1060 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1061 distribution's one) to use autoconf
1063 6. Fixed a few minor bugs
1065 Changes between versions 0.2e and 0.2f
1066 ======================================
1068 1. Added the creation of named pipes (fifos) in restore.
1070 2. Added the -N flag in the restore manual page.
1072 3. Added the file kernel.patch which contains the llseek() optimization
1073 patch for 1.2.x kernels.
1075 4. Fixed a bug in the restoration of symbolic links: owner and group were
1078 5. Integrated some changes from FreeBSD 2.2.
1080 6. Added a call to ftruncate() after restoring each file to restore
1081 correctly files ending by a hole.
1083 Changes between versions 0.2d and 0.2e
1084 ======================================
1086 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1087 with owner = root (0) and group = root/wheel/whatever (0).
1089 Changes between versions 0.2c and 0.2d
1090 ======================================
1092 1. Dump is now able to backup 2GB+ filesystems.
1094 2. Dump and restore can now be linked as static binaries.
1096 Changes between versions 0.2b and 0.2c
1097 ======================================
1099 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1101 Changes between versions 0.2a and 0.2b
1102 ======================================
1104 1. Really fixed the bug that I should have corrected in 0.2a.
1106 2. Enabled optimization again.
1108 Changes between versions 0.2 and 0.2a
1109 =====================================
1111 1. Disabled the optimization during compilation.
1113 Changes between versions 0.1 and 0.2
1114 ====================================
1116 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1117 the fs_type field (actually, I modified the file fstab.c to make it
1118 use the mntent functions).
1120 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1121 made by dump should be readable by the BSD restore and backups made
1122 by the BSD dump should be readable by restore. Unfortunately, this
1123 means that the dump format has changed between version 0.1 and version
1126 3. Dump is now able to backup a subtree, it is no longer limited to whole
1127 filesystems like the BSD version.
1129 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1132 Changes between versions 0.0 and 0.1
1133 ====================================
1135 1. Now create links rdump and rrestore during the `make install' step.
1137 2. Linux port specific bugs added to the manual pages
1139 3. Incorrect estimation of the number of tapes blocks fixed when doing
1140 incremental backups.
1142 4. Better ls-like format in restore in interactive mode.