1 $Id: CHANGES,v 1.214 2003/02/11 09:56:46 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 2. Fixed the configure script to only check for headers presence
11 instead of trying to compile them. This should fix issues
12 with old build environments. Thanks to Kari Lempiainen
13 <kari@funky.pp.fi> for reporting the bug.
15 3. Fixed restore to correctly ignore sockets when comparing
16 a dump (as socket cannot be properly restored anyway). Thanks
17 to <gunther@mit.edu> for reporting the bug.
19 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
20 =======================================================================
22 1. Added a note in the restore man page clarifying the question
23 'set the permissions on the current directory ?' asked by
24 restore at the end of treatment in -i and -x modes.
26 2. Fixed the endianess issues when reading compressed tapes.
27 Thanks to Dark Force <daq4th@users.sourceforge.net> for
28 reporting this bug and providing test cases.
30 3. Fixed the "ACL won't be dumped" warning message (which showed
31 an extra, unrelated error message). Thanks to Dragan Krnic
32 <dkrnic@lycos.com> for reporting this bug.
34 4. Made dump look first into /etc/mtab, then into /etc/fstab
35 when searching for filesystem paths. Also fixed some problems
36 caused by binding mounts or multiple block device mounts.
37 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
38 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
39 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
40 and #160305, Sourceforge bugs #618699 and #679832).
42 5. Made dump's -I option accept the value '0' meaning all the
43 read errors will be ignored. This can be useful when running
44 dump from unattended sessions (like cron jobs). Thanks to
45 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
47 6. Fixed the output of dump to indicate 'blocks' instead of
48 'tape blocks' in the various messages (blocks are always
49 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
50 made it clearly print the current blocksize at the start of
51 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
54 7. Made rmt understand long (largefiles) seeks.
56 8. Fixed build with very old versions of libext2fs, where
57 EXT2_FT_* constants were undefined.
59 9. Made dump accept the dumpdates path on the command line
60 (-D file option) instead of using only the hardcoded one.
61 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
64 10. Enabled rmt, LFS, readline, QFA options by default in
65 ./configure. Updated the configure process (new versions
66 of config.guess, config.sub etc).
68 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
69 =======================================================================
71 1. Changed dump to use fcntl(F_SETLK) style locking instead
72 of flock() when locking the dumpdates file. With the old
73 locking scheme, a local user having read rights on the
74 dumpdates file could be able to do a Denial of Service attack
75 on dump. In order to lock the dumpdates file with the new
76 scheme, the user would need to have write access on the file.
77 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
78 reporting the bug (originally a bugtraq post).
80 2. Fixed interactive 'ls' which caused spurious errors warnings
81 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
82 <jorgen_ostling@users.sourceforge.net> for reporting this
85 3. Fixed dump's estimate when dealing with sparse inodes.
87 4. Modified dump to allow setting a blocksize bigger than 32kB
88 (raised the limit to 1024kB), because newer hardware needs
89 this for proper operation. Thanks to Dirk Traenapp
90 <dtraenapp@users.sourceforge.net> for reporting this.
92 5. Fixed a bug causing Dump to stop and report an error if an
93 inode number in the exclude file was followed by some amount
94 of whitespace. Thanks to Jeffrey Sofferin
95 <sofferin@users.sourceforge.net> for reporting this bug.
97 6. Fixed a bug which caused restore, in some particular cases,
98 to ask some 'scary' questions and leave a bunch of RSTTMP
99 directories behind when restoring incremental tapes. Thanks
100 to Philippe Troin <phil@fifi.org> for reporting this bug and
101 providing the test cases.
103 7. Changed the wording when inodes are excluded from dump:
104 replaced 'Added inode 7 to exclude list' with
105 'Excluding inode 7 (resize inode) from dump', as suggested
106 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
108 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
109 ===================================================================
111 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
112 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
113 <eros@lamel.bo.cnr.it> for reporting the bug.
115 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
116 ===================================================================
118 1. Made dump print out the ext2 logical block number in case of a read
119 error, which can be given as an argument to debugfs' ncheck command,
120 in order to find out the affected inode. Added note about this
121 usage in dump's man page.
123 2. Fixed a problem in restore when reading tapes written on big
124 endian machines with very old versions of dump. The patch was
125 contributed by George Helffrich <george@geology.bristol.ac.uk>.
127 3. Fixed the tape length calculation when using large tapes
128 and compression. Thanks to Georg Lippold
129 <g_lippold@sourceforge.net> for reporting the bug.
131 4. Added a new set of examples in dump_on_cd_2 directory, based
132 on dump_on_cd examples but somewhat enhanced, supporting DVD
133 media, and localized in english and german. Thanks to
134 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
136 5. Made dump save 32 bit UID/GID. Internally, this was achieved
137 by switching from the old BSD inode format to the new BSD
138 inode format, which means that the tape format was changed.
139 However, since all restore versions out there should
140 transparently support both inode formats, the change should
141 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
144 6. Fixed a lot of warnings in the code shown when compiling
145 with 'gcc -W'. Thanks to Matthias Andree
146 <matthias.andree@stud.uni-dortmund.de> for reporting this.
148 7. Fixed a small markup bug in the dump man page. Thanks to
149 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
152 8. Rewrote entirely the man pages using the tmac.an macro
153 package (Linux man page format) instead of the original BSD
154 format. They should be now cleaner and easier to modify.
156 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
157 ==================================================================
159 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
160 non Linux clients were misinterpreted. The description of the problem
161 (incompatible numbering in Linux mtio opcodes) is documented at
162 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
163 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
164 providing an excellent, cross-platform replacement for rmt in his
167 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
168 which caused restore to fail when the inode maps were not entirely
169 contained on the first volume (when using really small volumes or
170 when dumping a huge number of inodes).
172 3. Fixed a problem in dump, where files in subdirectories of directories
173 that have the nodump flag set (or being excluded with -e/-E) were
174 sometimes incorrectly dumped (instead of being excluded). The original
175 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
176 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
177 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
178 help testing the patch.
180 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
181 which makes one able to pipe the output of dump, by the net, to
182 a remote CD-burner server.
184 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
185 and use GNU's symbolic syntax over rmt instead of numerical values
186 to assure multiple platform compatibility.
188 6. Documented the -d option in restore.
190 7. Added a -v (verbose) mode to dump. For now it just prints the number
191 of the inode being dumped, but this could evolve in future versions
192 to include interesting debugging output.
194 8. Added a -o flag to restore, which automatically restores the current
195 directory permissions in -i or -x mode, without asking the operator
196 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
197 and Peter Samuel <peters@e-smith.com>.
199 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
200 ====================================================================
202 1. Fixed a bug in the QFA feature which could have caused
203 a bad tape positionning on restore (causing extra delays in
204 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
206 2. Added a small note in the dump man page specifying that
207 there should be no white space between the option letter and
208 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
210 3. Made restore work with older versions of the readline library,
211 by checking in configure for several extended symbols. Restore
212 can now be compiled with a readline as old as the 2.0 release
213 (though it may be a good idea to upgrade it to a more recent
214 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
215 reporting the build failures.
217 4. Fixed a performance problem with the QFA file creation in
218 dump, which made unnecessary seeks on the tape slowing down
219 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
220 reporting this issue.
222 5. Removed the inclusion of some kernel headers in the dump
223 source, which prevented the compile in some kernel/glibc
224 headers/architecture combination. Thanks to Bdale Garbee
225 <bdale@gag.com> for reporting the bug.
227 6. Added the appropriate error message when dump fails to
228 open the output file for writing. Thanks to Amith Varghese
229 <amithv@yahoo.com> for reporting this bug.
231 7. Made restore able to understand large Solaris ufsdump tapes
232 (containing inodes bigger than 4194304). Sun have introduced
233 an "extension" to the dump tape format when dealing with
234 those inodes, which was uncorrectly handled by Linux restore.
235 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
236 providing a test case.
238 8. Added the -m parameter to dump which optimises the output for
239 inodes having been changed but not modified since the last dump
240 ('changed' and 'modified' have the meaning defined in stat(2)).
241 For those inodes, dump will save only the metadata, instead of
242 saving the entire inode contents. Inodes which are either
243 directories or have been modified since the last dump are saved
244 in a regular way. Uses of this flag must be consistent, meaning
245 that either every dump in an incremental dump set have the flag,
248 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
249 =======================================================================
251 1. Fixed behaviour of dump when exceeding resource limits
254 2. Added the -L flag to restore to allow the user to specify a
255 maximal allowed number of miscompares when using restore
256 with the -C option to check the backup.
258 3. Detailed the manual entry for the -N option of restore.
260 4. Added the -a flag to restore to make able doing unattended
261 restores in -i or -x mode (automatically walks through the
262 multiple dump volumes).
264 5. Extended the QFA mode to work with local files and/or
265 remote tapes and files. This way, restore can know in advance
266 the tape number and the offset for the inodes to extract and
267 can minimize the extraction time by seeking directly to the
270 6. Added the -A <archive> option to both dump and restore,
271 which makes dump to archive a dump table-of-contents in
272 the specified file to be used by restore to determine
273 whether a file is in the dump file that is being restored.
274 (the archive file syntax is also compatible with the
275 Solaris ufsdump generated one).
277 7. Small fix in restore making it able to read some (broken ?)
278 Solaris ufsdump tapes.
280 8. Fixed dump to correctly recognise the root filesystem when using
281 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
282 for reporting this bug.
284 9. Added the -P <file> option to restore to create a
285 Quick File Access file from an already made dump. Patch
286 contributed by Uwe Gohlke <uwe@ugsoft.de>.
288 10. Made restore compile and run on Solaris, making it a
289 possible replacement for the standard ufsrestore. Port was
290 contributed by Uwe Gohlke <uwe@ugsoft.de>.
292 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
293 =====================================================================
295 1. Added a set of backup scripts from Eugenio Diaz
296 <getnito@yahoo.com> in the examples section. It features
297 automatic (cron based) full and incremental dumping of
298 several filesystems on a separate filesystem.
300 2. Fixed a off-by-one miscalculation which disabled dumping
301 a one letter subdirectory. Thanks to Chris Siebenmann
302 <cks@utcc.utoronto.ca> for reporting the bug.
304 3. Fixed several restore bugs occuring when trying to
305 restore some missing files on the tape. Thanks to Chris
306 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
308 4. Fixed --with-ldopts configure argument passing, installing from
309 a separate object directory, makefile cleanups contributed
310 by <splite@purdue.edu>.
312 5. Fix a bug which could caused, in some conditions, the highest
313 number inode of a filesystem, to not be dumped. Many thanks
314 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
317 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
318 =======================================================================
320 1. Added a mini howto from Patrick Walsh in the examples
323 2. Minor man pages syntax corrections. Thanks to
324 Chris Danis <screechco@home.com> for reporting the bugs.
326 3. Added a script from David B. Peterson <dave@toppledwagon.com>
327 to the examples section. It features dumping several
328 filesystems to a remote tape drive upon ssh.
330 4. Added a patch provided by Richard Jones <rich@annexia.org>
331 which allows BRADEMAX (number of read errors tolerated by
332 dump) to be adjusted using the -I option.
334 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
335 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
338 6. Add the -l option to restore to specify if, when doing a
339 remote restore, the file used to access the backup is a
340 regular file (the defaults being a tape drive). Restore needs
341 to know this information when reading a remote compressed
342 dump. Previously, this information was autodetected, but
343 the autodetection code fails (with ioctl: Inappropriate ioctl
344 for device) when using a non Linux remote box. Thanks to
345 many users and especially to Eros Albertazzi
346 <eros@lamel.bo.cnr.it> for reporting this.
348 7. Found a workaroung for the dump deadlock problem (3 childs
349 stuck in pause(), father in read()). The workaround seems
350 to work for me and several beta-testers. If it doesn't work
351 for you, please report back.
353 8. Updated the RPM spec file (BuildPrereq, URL etc).
355 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
356 ========================================================================
358 1. Fixed the permissions of a newly created QFA file by dump.
360 2. Cleaned up the source of dump (the external variables
361 definition was a complete mess, making possible to have
364 3. Fixed restore to use the full tape volume path when doing
365 a compare (since it changes the working directory to the
366 filesystem being compared in the process).
368 4. Added the -q option to dump which makes dump abort
369 whenever it needs operator attention. It should be
370 easier to use dump in scripts with this option.
372 5. Detect the use of incompatible options to dump and
373 refuse them (like -a and -B options together).
375 6. Added bzip2 compression to dump/restore (use option -j level
376 to select it). Note that you will need a restore version
377 >= 0.4b24 in order to restore a bzip2 compressed dump.
378 The same warning as for the zlib compression applies:
379 the tape format of a bzip2 dump is not compatible with the
380 original BSD tape format.
382 7. Fixed a overflow problem in dump corrupting the dump when
383 very large files were encountered. Thanks to Vladimir Nadvornik
386 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
387 kernel buffer/page cache before starting the dump, helping
388 a bit those who use dump on mounted filesystems. Thanks to
389 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
392 9. Updated the RPM spec file following the RedHat changes
393 (dynamically linked binaries now in /usr/sbin etc).
395 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
396 which enables restore to recognise multi volume compressed dumps
397 done on CD. Included his scripts for dump (which pipe the dump
398 output directly into cdrecord) and restore. There is now
399 possible to to backups to CD on the fly!
401 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
402 ===================================================================
404 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
405 Marcus Meissner (Caldera International Security Dept.).
407 2. Implement the Sun rmt extended protocol. Patch submitted
408 by Ian Gordon <iangordon@users.sourceforge.net>.
410 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
411 the linux kernel header. This ensures that dump always has the
412 latest version of this file. Patch submitted by
413 Andreas Dilger <adilger@turbolinux.com>.
415 4. Report any filesystem present in either /etc/fstab with a
416 positive passno or /etc/dumpdates in dump -w output.
417 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
419 5. Fixed the looping problem in dump introduced in the
422 6. Changed the -B option of dump to limit the size of
423 _compressed_ output per volume if compression is on.
424 Patch contributed by Helmut Jarausch
425 <jarausch@igpm.rwth-aachen.de>. Note however that, since
426 it is impossible to predict the size of the compressed
427 data before doing the compression, the -B limit is a bit
430 7. Fixed a bug in reading the operator typed file/tape path for
431 the next volume in restore.
433 8. Implemented a "-F script" option to restore which permits the
434 user to specify a script which will be launched at the
435 beginning of each tape, useful for automatic programming of
436 tape changers for example. See the restore man page for the
437 script parameters and return codes.
439 9. Small fix for the QFA routines provided by Uwe Gohlke
440 <uwe@ugsoft.de>, and some recommendations for QFA uses in
443 10. Fixed the multivolume restoring where making a mistake
444 to the 'Mount next tape' prompt caused several blocks to
447 11. Enhanced the -e option of dump to take as a parameter a
448 comma separated list of inode numbers.
450 12. Added the -E option to dump which specify a file containing
451 inode numbers to exclude from the dump.
453 13. Fixed the compressed multi-volume dump + restore.
455 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
456 ==================================================================
458 1. Made dump report the number of blocks dumped per volume.
459 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
461 2. Fix a bug in dump where waiting too much at the 'change volumes'
462 question caused the volume to be rewritten. Thanks to
463 Graham Reed <greed@users.sourceforge.net> for reporting the
464 bug and providing a patch.
466 3. Added a compression option to dump, contributed by Jerry
467 Peters <gapeters@worldnet.att.net>.
469 WARNING: the compression works only on files/pipes and
470 drives supporting variable block size.
472 WARNING2: tapes created using the compression feature are
473 incompatible with the standard dump tape format, and a
474 version of restore >= 0.4b22 will be needed for extraction.
476 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
477 architectures. Thanks to Paul Slootman <paul@debian.org> for
478 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
481 5. Many cleanups (CPP defines, const char warnings, check of
482 ext2fs COMPAT flags, time_t cleanups, added libext2 version
483 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
485 6. Made --prefix option work in configure. All the install path
486 are now based on the configure parameters.
488 7. Added the Quick File Access mode in dump/restore, contributed
489 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
490 a file tape position for each inode, and this file is used by
491 restore (if called with parameter Q and the filename)
492 to directly position the tape at the file restore is currently
493 working on. This saves hours when restoring single files from
494 large backups, saves the tapes and the drive's head. Use
495 --enable-qfa option of configure to compile in the QFA support.
497 8. Added the possibility to dump several files and directories
498 in a single invocation of dump. Thanks to Uwe Gohlke
499 <uwe@ugsoft.de> for implementing this option.
501 9. Fixed the dumping and restoring of files with holes
502 (files ending with holes didn't get dumped or restored
505 10. Fixed a socket descriptor leak in dump, which leaved opened
506 3 file descriptors per dump process (and there is one dump
509 11. Fixed dump large file system support, by explicit use of
510 open64/lseek64/etc functions (explicit use needed because
511 e2fsprogs libraries don't behave well when compiled with
512 FILE_OFFSET_BITS=64).
514 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
515 ======================================================================
517 1. Fixed some bugs in the dump -w|-W logic introduced by
518 the previous version. Thanks to Andreas Dilger
519 <adilger@turbolinux.com> for his help on this one.
521 2. Fixed again a compilation problem when using old e2fs
522 headers (filesystem label related). Thanks to many users
523 who reported this stupid error.
525 3. Fixed a build problem on old lib5 based systems dealing with
526 _PATH_MNTTAB being undefined. Thanks to John Adams
527 <johna@onevista.com> for reporting the bug.
529 4. Improved the error detection in restore when creating
530 the temporary files in TMPDIR. Now restore will corectly
531 report a 'no space left on device' error instead of
532 strange errors which could imply an archive corruption.
533 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
534 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
536 5. Added the throughput information in dump's progression
537 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
540 6. Use libext2fs's inode scanning routines, which are particularly
541 robust when dealing with errors (especially when having some
542 bad blocks on the filesystem). Many thanks to Colin
543 <colin@horizon.com> for reporting the bug, investigating
544 the issues, finding a workaround, writing the patch and
545 fully testing it... (of course, if this patch will break
546 anything, he is to blame for :-)).
548 7. Made dump and restore LFS aware. Dump can dump a filesystem
549 containing large files, generate a large file on output and
550 restore can restore them. This feature is not enabled by
551 default, use --enable-largefile option of configure to enable
552 it (you need to have a LFS aware glibc though). Thanks to
553 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
554 and to Theodore T'so <tytso@valinux.com> for his always
557 8. Made dump ask upon a tape write error if it should rewrite
558 the current volume (assume this is a bad tape) or if it should
559 assume an end-of-tape condition (useful for tape drives which
560 behaves badly at the end of the tape). Thanks to Andreas
561 Dilger <adilger@turbolinux.com> for the suggestion.
563 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
564 =======================================================================
566 1. Fixed a small compilation problem due to a change
567 in the definintion of the struct sigaction in
568 glibc 2.0 / libc5. Thanks to Gunther Schlegel
569 <schlegel@riege.de> for reporting the bug and to
570 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
573 2. Modified the label and uuid analysis in order to be
574 self-contained (independant of kernel/libc headers). This
575 should fix the compile with older kernel/libc headers and
576 will preserve the functionality. Thanks to Bernhard Erdmann
577 <bernhard.erdmann@gmx.de> for reporting the bug.
579 3. The 'exclude inode' option, if applied to a directory
580 inode, excludes now correctly all the leaves of this
581 directory. Thanks to John R. Dennison
582 <gerdesas@users.sourceforge.net> for reporting the bug.
584 4. Fixed the '-e' option to disable the possibility
585 to exclude the root inode (which causes the dumps to
586 be unreadable by restore). Prevented array overflow
587 when multiple -e options are used.
589 5. Fixed dump to correctly interpret a filesystem argument
590 which is a mountpoint and it is not an absolute path
591 (as specified in the fstab). Thanks to Bernhard R. Erdmann
592 <be@berdmann.de> for reporting the bug.
594 6. Made dump able to backup files larger than 2 GB. Note that
595 dump still doesn't cope with files larger than 4 GB.
597 7. Restore the real uid and gid before invoking an external
598 RSH program (big hole when dump or restore is suid root!).
600 8. Get the values for the default tape device and the location
601 of fstab file from the system headers. Thanks to
602 Andreas Dilger <adilger@turbolinux.com> for the patch.
604 9. Made dump -w|-W report all recognized filesystems
605 present in either /etc/fstab or /etc/dumpdates, and present
606 the list in the same order as in fstab file. Thanks
607 to Andreas Dilger <adilger@turbolinux.com> for the patch.
609 10. Made dump's -a (automatic end of tape detection) the
610 default. Specifying one of -B, -c, -d or -s options will
611 override the EOM detection. Thanks to Andreas Dilger
612 <adilger@turbolinux.com> for the patch.
614 11. Save the ext2 filesystem label into the dump volume label.
615 Specifying a volume label on the command line (-L option)
616 will override this feature. Thanks to Andreas Dilger
617 <adilger@turbolinux.com> for the patch.
619 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
620 =====================================================================
622 1. Fixed the signal handling in dump (which I broke in 0.4b17)
623 which was causing several strange dump failures (dump
624 hanged or segmentation faults etc).
626 2. Specified the default blocksize in the dump man page.
628 3. Changed two info messages of restore to be written on stdout
629 instead of stderr in order to leave stderr only for errors
630 or warnings. Thanks to Stephen Atwell
631 <satwell@urbana.css.mot.com> for the suggestion.
633 4. Corrected an off by one calculation which prevented
634 dumping a directory having a 1 character name length.
635 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
636 for reporting the bug.
638 5. Reinforce dump in order to allow only level 0 and no
639 -u option when dumping a subdirectory, and document
640 this in the man page. Thanks to Bernhard Erdmann
641 <bernhard.erdmann@gmx.de> for reporting the bug.
643 6. Fixed a small harmless bug in restore which caused
644 in some conditions hard links to be restored several
645 times (therefore generation some warning messages).
646 Thanks to Kenneth Porter <shiva@well.com> for
649 7. Updated the RPM spec file to the latest RedHat version,
650 providing FHS packaging and other cosmetic changes.
651 You will need RPM version >= 3.0.5 in order to build the RPM.
653 8. Updated the configure script to check for libtermcap
654 before checking for libreadline, because we need this
655 library in order to compile the readline support.
657 9. Made dump understand the LABEL= and UUID= notation
658 both in /etc/fstab and on the command line. Note that
659 you will need the /proc filesystem in order to use
660 these notations. Thanks to Erik Troan <ewt@redhat.com>
661 for providing the patch.
663 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
664 ===================================================================
666 1. Fixed a potential buffer overflow in restore. Thanks
667 to Stan Bubrouski <satan@fastdial.net> for reporting
670 2. Fixed a readline-related bug which prevented
671 'cat DUMPFILE | restore -i -f -' from working. Thanks
672 to Charles Karney <karney@users.sourceforge.net>
675 3. Changed a few "panic" into "exit", causing restore to
676 be more stable against some attacks (like the last one
677 reported on Bugtraq, although the last version of restore
678 was not vulnerable - just dumped core). Thanks to
679 Andreas Hasenack <andreas@conectiva.com.br> for reporting
682 4. Removed the suid-root bit on dump and restore in the
683 default build (and generated RPMs). It should be safer
684 now. Users who need the suid-root capabilities in order
685 to make network backups should read first the man page
686 of dump and enable the suid bit manually.
688 5. Added -ltermcap to the compile parameters for restore
689 when using readline support, in order to make the compile
690 process work on some systems (SuSE ?). Thanks to
691 Patrik Schindler <poc@pocnet.net> for reporting the bug.
693 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
694 ====================================================================
696 1. The -F script is called now *only* at the end of a tape,
697 not at the end of the dump. Thanks to Christian Haul
698 <haul@informatik.tu-darmstadt.de> for the bug report.
700 Normally, the device name and the current volume number
701 are passed on the command line of the script. However,
702 if you want the old style script (with no arguments
703 passed on the command line), you can enable it in
704 configure with the --enable-oldstylefscript.
706 2. Use posix signal handling to preserve dump functionality
707 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
710 3. Made the exit code of restore in case of a 'C'ompare
711 command reflect the result of the compare. An exit status
712 of 0 means the dump archive is correct, 1 means tape errors,
713 2 means that some files were modified. Thanks to Kenneth Porter
714 <shiva@well.com> for the suggestion.
716 4. Made (finally) quotes work in interactive restore.
718 5. Small fixes in order to allow dump to compile with a
719 really old e2fsprogs version. Thanks to Ian Zimmerman
720 <itz@speakeasy.org> for the bug report.
722 6. Add GNU readline capabilities to interactive restore.
723 Use configure's --enable-readline flag to enable this feature.
724 Thanks to Patrik Schindler <poc@pocnet.net> for the
727 7. Do the compare on the fly in restore 'C' mode (this will
728 allow not to exhaust the available /tmp space when
729 ccmparing large files). Thanks to Kenneth Porter
730 <shiva@well.com> for the suggestion.
732 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
733 ====================================================================
735 1. Fixed some several duplicate 'DUMP: DUMP:' in the
738 2. Corrected the estimation of blocks to dump. Note that
739 this is still wrong for large directory inodes, due
740 to the size of a BSD directory entry on the tape
741 (estimation is based on the size of the ext2 directory,
742 which is smaller because it doesn't contain the null
743 character at the end).
745 3. Made dump report the total number of MB written to
746 tape at the end of run. Thanks to W. Reilly Cooley
747 <wcooley@nakedape.cc> for the patch.
749 4. Added the -X option to restore giving the possibility
750 to read the names of the files to be extracted/listed
751 from a text file (in addition of the command line).
752 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
755 5. Added the device name and the current volume number
756 as arguments to the end of tape script (dump -F option).
758 6. Made the multi-volume dump work again (after having
759 broken it in 0.4b15).
761 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
762 ===================================================================
764 1. Added a prompt command in interactive restore mode. Thanks
765 to Andreas Dilger <adilger@home.com> for the patch.
767 2. Fixed a buffer overflow problem in dump (caused by
768 not checking the size of the filesystem parameter).
769 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
770 reporting this on Bugtraq (and to several dump users
771 who forwarded me his mail).
773 3. Added the '-F script' option to dump in order to
774 launch a script at the end of each tape (to be used
775 with a tape changer, or to notify the sysadmin by
778 4. Fixed a bug in restore compare code caused by the changes
781 5. Fixed the treatment of options using the old BSD syntax
782 in both dump and restore.
784 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
785 =======================================================================
787 1. Fixed a bug in dump which may have caused invalid deleted
788 directories to be dumped out if they were deleted after the
789 mapping pass. This could occure on active filesystem and lead
790 to heap corruption (causing dump malfunction in many possible ways).
791 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
792 bug and submitting the patch.
794 2. Fixed the handling of the filesystem parameter in dump. A
795 '/mnt/disk' parameter caused the disk contents to be dumped,
796 but a '/mnt/disk/' parameter caused the mountpoint directory
797 to be dumped (generally an empty directory).
799 3. Improved the output of dump in order to tell which directory
800 it is currently dumping (when dumping a subtree).
802 4. Added the '-e' exclude inode option to dump. Thanks to
803 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
805 5. Added a REPORTING-BUGS file in order to provide a guide
806 on how to correctly report a bug in dump/restore.
808 6. Fixed a restore problem when restoring a hard link to an inode
809 having the immutable or append-only attribute set. Thanks to
810 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
812 7. Fixed a compatibility problem between dumps made on little
813 endian machines (the format was correct) and big endian
814 machines (the format was incorrect). This fix break the
815 compatibility with the older dumps made on big endian
816 machines (sparc, m86k, ppc etc). For the first time in
817 linux dump's history (I believe), the dumps made by *BSD,
818 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
819 so interchangeable. Thanks to Rob Cermak
820 <cermak@ahab.rutgers.edu> for submitting the bug and
821 helping me test the fix.
823 8. Fixed the way dump reports the remaining percent/time, if
824 the number of blocks actually dumped exceeds the estimated
825 number of blocks. Thanks to Jean-Paul van der Jagt
826 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
828 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
829 ======================================================================
831 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
832 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
834 2. Removed the requirement to build the RPM as root from the
835 spec file. Thanks to Christian Weisgerber
836 <naddy@mips.rhein-neckar.de> for submitting this.
838 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
839 causing dump to block when dumping really huge filesystems.
840 Many thanks to Patrik Schindler <poc@pocnet.net> for
841 helping me find this bug.
843 4. Fixed the treatment for an interrupt signal when dump access
844 the remote tape through RSH. Thanks to Christian Weisgerber
845 <naddy@mips.rhein-neckar.de> for providing the patch.
847 5. Fixed a bug which was causing dump/restore to display
848 garbage characters instead of the remote host name.
850 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
851 =====================================================================
853 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
854 <kukuk@suse.de> for submitting the patch.
856 2. Fix for the exit code when using the size estimate option of
857 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
858 submitting the patch.
860 3. Handle EINTR in atomical reads/writes in dump, which was causing
861 dump to fail on some systems. Thanks to Eric Jergensen
862 <eric@dvns.com> for reporting the bug and submitting the patch.
864 4. Handle more than 16 characters for the device names in dumpdates.
865 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
866 tracking down the problem and proposing the solution.
868 5. Fixed a bug in dump which prevented the creation of the
869 dumpdates file when doing a 0-level dump without already
870 having a dumpdates file. Thanks to Patrik Schindler
871 <poc@pocnet.net> for reporting the bug.
873 6. Changed the way dump 'S' flag reports the size estimate
874 from number of blocks into bytes (making it compatible
875 with the Solaris version, and simplifying things for
876 amanda users). Thanks to Jason L Tibbitts III
877 <tibbs@math.uh.edu> for reporting the bug.
879 7. Fixed a compatibility problem in linux/alpha dump tape format.
880 Now the linux/alpha dump are (again) compatible with the
881 other dump formats. But this breaks compatibility with
882 older dumps made on alpha. Thanks to Mike Tibor
883 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
885 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
886 ======================================================================
888 1. Added a '--enable-kerberos' to configure.
890 2. Added a 'S' option to dump which determines the amount of space
891 that is needed to perform the dump without actually doing it, similar
892 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
893 <cermak@ahab.rutgers.edu>.
895 3. Added a 'M' multi-volume option to dump and restore which enables
896 dumping to multiple files (useful when dumping to an ext2
897 partition to make several dump files in order to bypass the 2GB
898 file size limitation). The 'f' argument is treated as a prefix and
899 the output files will be named <prefix>001, <prefix>002 etc. With
900 the 'M' flag, restore automatically selects the right file without
901 asking to enter a new tape each time.
903 4. Fixed a memory leak which was causing dump to grow very big
904 (270MB when dumping a 10GB filesystem...). Thanks to Jason
905 Fearon <jasonf@netrider.org.au> for reporting the bug.
907 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
908 ======================================================================
910 1. Make configure test if the system glob routines support
911 extended syntax (ALTDIRFUNC). If not, use the internal glob
912 routines instead of system ones. Thanks to Bernhard Sadlowski
913 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
914 and helping me resolve this and other minor libc5 compiling
917 2. Fix a problem when dumping a ext2fs with the 'filetype'
918 feature enabled. Thanks to Patrick J. LoPresti
919 <patl@cag.lcs.mit.edu> for reporting the bug and to
920 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
922 3. Made the nodump flag work on directories. A directory which
923 has the nodump flag gets never dumped, regardless of its
926 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
927 which allows dump on an active ext3 filesystem. However, this
928 is a "quick and dirty" patch which enables backup of an ext3
929 filesystem through the ext2 compatibility (by ignoring the
930 NEEDS_RECOVERY bit). The journal file is not recognized and
931 it is dumped (it should not).
933 5. Test the superblock compatibility flags when dumping, in order
934 to be sure that we know how to deal with specific features.
936 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
937 ====================================================================
939 1. Use lchown instead of chown, fixing a possible security problem
940 when restoring symlinks (a malicious user could use this
941 to deliberately corrupt the ownership of important system files).
942 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
943 this and providing the patch.
945 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
946 ====================================================================
948 1. Put dump sources under CVS, added Id tags in all files so
949 one can use 'ident' on binary files.
951 2. Added the dump/restore version in the usage text so one can
952 easily verify the version he is using.
954 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
955 a va_start/va_end problem on linux-ppc (always call va_start
956 va_end in pairs each time we use a vararg function).
958 4. Added again the DT_* constants because old libc does not
959 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
960 for submitting the bug report.
962 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
963 this should enable dumping big (huge) filesystems.
965 6. Added the RSH environment variable in order to be able to
966 use a rsh replacement like ssh when doing remote backups (and
967 bypass the security limitations of rcmd). Now you can do remote
968 backups without being root (or making dump setuid root).
970 7. Modified again the way dumpdates works. For incremental dumps,
971 we need to read dumpdates even if we are not using 'u' option.
972 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
975 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
976 ===================================================================
978 1. Removed the 'k' flag from the restore 'about' text if kerberos
981 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
982 to them (fsetflags takes a char*, setflags an open fd!).
984 3. (f)setflags is called only if the flags aren't empty. If the
985 file is a special file, a warning is printed, because changing
986 flags implies opening the device. Normally, a special file
987 should not have any flag... (Debian bug #29775, patch provided
988 by Abhijit Dasgupta <abhijit@ans.net>).
990 4. Made possible to dump a file system not mentioned in /etc/fstab.
991 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
993 5. Changed the default behaviour to not create dumpdates
994 unless 'u' option is specified. Removed the old "debian-patch"
995 which provided the same thing. (Debian bug #38136, #33818).
997 6. Removed all those dump*announce, since they were getting old...
999 7. Added warning messages if dumpdates does not exist and
1000 when an inferior level dump does not exist (except for a level 0
1003 8. Debugged the glob calls in interactive mode: restore used a
1004 dirent struct which was different from the /usr/include/dirent.h
1005 one (this used to work, is it a glibc2 change?), so none of the
1006 compat glob (which used /usr/include/dirent.h) or the system glob
1007 worked. Restore use now the system dirent (and the system
1008 DT_* constants), which are compatible with BSD ones.
1010 9. Added a configure flag (--with-dumpdatespath) to specify
1011 the location of dumpdates file. By default, it is
1014 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1015 included the current date/version in man pages.
1017 11. Corrected the estimation of remaining time when
1018 the operator doesn't change the tapes quickly enough. This
1019 was an old bug, I thought I corrected it, and discovered
1020 that in fact it was corrected in two different places, so
1021 the results canceled each other...
1023 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1024 ===================================================================
1026 1. Integrated multiple patches from RedHat, Debian and SuSE:
1028 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1030 - several fixes in the man pages.
1031 - update the default tape device to /dev/st0.
1032 - many updates for Linux Alpha (byte ordering, size_t etc).
1034 - use environment variable for TMPDIR (instead of /tmp).
1035 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1036 - workaround egcs bug (RedHat bugs #4281 and #2989).
1037 - wire $(OPT) throughout Makefile's.
1039 2. Upgrade the dump revision to 1, making possible to dump filesystems
1040 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1042 3. Fix some compile warnings, prototype all functions.
1044 4. Use glibc err/glob instead of internal compatibility
1045 routines (only if available).
1047 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1048 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1050 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1051 ======================================================================
1053 1. Integrated the changes from FreeBSD-3.1-RELEASE
1054 (mostly bug fixes, buffer overruns, dump has now an "automatic
1055 tape length calculation" flag, dump/restore can use kerberos now
1056 (this is NOT tested), use environment variables for TAPE and
1059 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1061 3. Corrected a bug in restore when using 'C' option with multi-volumes
1062 tapes (files splited accros two tapes give "size changed" errors
1065 4. Corrected the long standing bug when dumping multiple tapes.
1066 This works for me, needs further testing.
1068 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1069 ====================================================================
1071 1. Dump now runs correctly on kernels 2.1.x
1072 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1074 Changes between versions 0.4b2 and 0.4b3
1075 ========================================
1077 1. Use realpath() if available
1079 2. Report statistics
1081 Changes between versions 0.4b1 and 0.4b2
1082 ========================================
1084 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1087 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1089 3. Fixed configure.in to correctly handle --enable-rmt
1091 Changes between versions 0.3 and 0.4b1
1092 ======================================
1094 1. Integrated the changes from 4.4BSD-Lite2
1096 2. Integrated the patches from Debian and Red Hat
1098 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1100 4. Changed dump to use the Ext2fs library to get block addresses. This
1101 should solve the endianness problem on SparcLinux.
1103 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1104 distribution's one) to use autoconf
1106 6. Fixed a few minor bugs
1108 Changes between versions 0.2e and 0.2f
1109 ======================================
1111 1. Added the creation of named pipes (fifos) in restore.
1113 2. Added the -N flag in the restore manual page.
1115 3. Added the file kernel.patch which contains the llseek() optimization
1116 patch for 1.2.x kernels.
1118 4. Fixed a bug in the restoration of symbolic links: owner and group were
1121 5. Integrated some changes from FreeBSD 2.2.
1123 6. Added a call to ftruncate() after restoring each file to restore
1124 correctly files ending by a hole.
1126 Changes between versions 0.2d and 0.2e
1127 ======================================
1129 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1130 with owner = root (0) and group = root/wheel/whatever (0).
1132 Changes between versions 0.2c and 0.2d
1133 ======================================
1135 1. Dump is now able to backup 2GB+ filesystems.
1137 2. Dump and restore can now be linked as static binaries.
1139 Changes between versions 0.2b and 0.2c
1140 ======================================
1142 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1144 Changes between versions 0.2a and 0.2b
1145 ======================================
1147 1. Really fixed the bug that I should have corrected in 0.2a.
1149 2. Enabled optimization again.
1151 Changes between versions 0.2 and 0.2a
1152 =====================================
1154 1. Disabled the optimization during compilation.
1156 Changes between versions 0.1 and 0.2
1157 ====================================
1159 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1160 the fs_type field (actually, I modified the file fstab.c to make it
1161 use the mntent functions).
1163 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1164 made by dump should be readable by the BSD restore and backups made
1165 by the BSD dump should be readable by restore. Unfortunately, this
1166 means that the dump format has changed between version 0.1 and version
1169 3. Dump is now able to backup a subtree, it is no longer limited to whole
1170 filesystems like the BSD version.
1172 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1175 Changes between versions 0.0 and 0.1
1176 ====================================
1178 1. Now create links rdump and rrestore during the `make install' step.
1180 2. Linux port specific bugs added to the manual pages
1182 3. Incorrect estimation of the number of tapes blocks fixed when doing
1183 incremental backups.
1185 4. Better ls-like format in restore in interactive mode.