1 $Id: CHANGES,v 1.184 2002/07/01 11:57:11 stelian Exp $
3 Changes between versions 0.4b29 and 0.4b30 (released ????????????????)
4 ======================================================================
6 1. Made dump print out the ext2 logical block number in case of a read
7 error, which can be given as an argument to debugfs' ncheck command,
8 in order to find out the affected inode. Added note about this
9 usage in dump's man page.
11 2. Fixed a problem in restore when reading tapes written on big
12 endian machines with very old versions of dump. The patch was
13 contributed by George Helffrich <george@geology.bristol.ac.uk>.
15 3. Fixed the tape length calculation when using large tapes
16 and compression. Thanks to Georg Lippold
17 <g_lippold@sourceforge.net> for reporting the bug.
19 4. Added a new set of examples in dump_on_cd_2 directory, based
20 on dump_on_cd examples but somewhat enhanced, supporting DVD
21 media, and localized in english and german. Thanks to
22 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
24 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
25 ==================================================================
27 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
28 non Linux clients were misinterpreted. The description of the problem
29 (incompatible numbering in Linux mtio opcodes) is documented at
30 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
31 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
32 providing an excellent, cross-platform replacement for rmt in his
35 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
36 which caused restore to fail when the inode maps were not entirely
37 contained on the first volume (when using really small volumes or
38 when dumping a huge number of inodes).
40 3. Fixed a problem in dump, where files in subdirectories of directories
41 that have the nodump flag set (or being excluded with -e/-E) were
42 sometimes incorrectly dumped (instead of being excluded). The original
43 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
44 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
45 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
46 help testing the patch.
48 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
49 which makes one able to pipe the output of dump, by the net, to
50 a remote CD-burner server.
52 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
53 and use GNU's symbolic syntax over rmt instead of numerical values
54 to assure multiple platform compatibility.
56 6. Documented the -d option in restore.
58 7. Added a -v (verbose) mode to dump. For now it just prints the number
59 of the inode being dumped, but this could evolve in future versions
60 to include interesting debugging output.
62 8. Added a -o flag to restore, which automatically restores the current
63 directory permissions in -i or -x mode, without asking the operator
64 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
65 and Peter Samuel <peters@e-smith.com>.
67 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
68 ====================================================================
70 1. Fixed a bug in the QFA feature which could have caused
71 a bad tape positionning on restore (causing extra delays in
72 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
74 2. Added a small note in the dump man page specifying that
75 there should be no white space between the option letter and
76 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
78 3. Made restore work with older versions of the readline library,
79 by checking in configure for several extended symbols. Restore
80 can now be compiled with a readline as old as the 2.0 release
81 (though it may be a good idea to upgrade it to a more recent
82 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
83 reporting the build failures.
85 4. Fixed a performance problem with the QFA file creation in
86 dump, which made unnecessary seeks on the tape slowing down
87 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
90 5. Removed the inclusion of some kernel headers in the dump
91 source, which prevented the compile in some kernel/glibc
92 headers/architecture combination. Thanks to Bdale Garbee
93 <bdale@gag.com> for reporting the bug.
95 6. Added the appropriate error message when dump fails to
96 open the output file for writing. Thanks to Amith Varghese
97 <amithv@yahoo.com> for reporting this bug.
99 7. Made restore able to understand large Solaris ufsdump tapes
100 (containing inodes bigger than 4194304). Sun have introduced
101 an "extension" to the dump tape format when dealing with
102 those inodes, which was uncorrectly handled by Linux restore.
103 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
104 providing a test case.
106 8. Added the -m parameter to dump which optimises the output for
107 inodes having been changed but not modified since the last dump
108 ('changed' and 'modified' have the meaning defined in stat(2)).
109 For those inodes, dump will save only the metadata, instead of
110 saving the entire inode contents. Inodes which are either
111 directories or have been modified since the last dump are saved
112 in a regular way. Uses of this flag must be consistent, meaning
113 that either every dump in an incremental dump set have the flag,
116 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
117 =======================================================================
119 1. Fixed behaviour of dump when exceeding resource limits
122 2. Added the -L flag to restore to allow the user to specify a
123 maximal allowed number of miscompares when using restore
124 with the -C option to check the backup.
126 3. Detailed the manual entry for the -N option of restore.
128 4. Added the -a flag to restore to make able doing unattended
129 restores in -i or -x mode (automatically walks through the
130 multiple dump volumes).
132 5. Extended the QFA mode to work with local files and/or
133 remote tapes and files. This way, restore can know in advance
134 the tape number and the offset for the inodes to extract and
135 can minimize the extraction time by seeking directly to the
138 6. Added the -A <archive> option to both dump and restore,
139 which makes dump to archive a dump table-of-contents in
140 the specified file to be used by restore to determine
141 whether a file is in the dump file that is being restored.
142 (the archive file syntax is also compatible with the
143 Solaris ufsdump generated one).
145 7. Small fix in restore making it able to read some (broken ?)
146 Solaris ufsdump tapes.
148 8. Fixed dump to correctly recognise the root filesystem when using
149 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
150 for reporting this bug.
152 9. Added the -P <file> option to restore to create a
153 Quick File Access file from an already made dump. Patch
154 contributed by Uwe Gohlke <uwe@ugsoft.de>.
156 10. Made restore compile and run on Solaris, making it a
157 possible replacement for the standard ufsrestore. Port was
158 contributed by Uwe Gohlke <uwe@ugsoft.de>.
160 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
161 =====================================================================
163 1. Added a set of backup scripts from Eugenio Diaz
164 <getnito@yahoo.com> in the examples section. It features
165 automatic (cron based) full and incremental dumping of
166 several filesystems on a separate filesystem.
168 2. Fixed a off-by-one miscalculation which disabled dumping
169 a one letter subdirectory. Thanks to Chris Siebenmann
170 <cks@utcc.utoronto.ca> for reporting the bug.
172 3. Fixed several restore bugs occuring when trying to
173 restore some missing files on the tape. Thanks to Chris
174 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
176 4. Fixed --with-ldopts configure argument passing, installing from
177 a separate object directory, makefile cleanups contributed
178 by <splite@purdue.edu>.
180 5. Fix a bug which could caused, in some conditions, the highest
181 number inode of a filesystem, to not be dumped. Many thanks
182 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
185 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
186 =======================================================================
188 1. Added a mini howto from Patrick Walsh in the examples
191 2. Minor man pages syntax corrections. Thanks to
192 Chris Danis <screechco@home.com> for reporting the bugs.
194 3. Added a script from David B. Peterson <dave@toppledwagon.com>
195 to the examples section. It features dumping several
196 filesystems to a remote tape drive upon ssh.
198 4. Added a patch provided by Richard Jones <rich@annexia.org>
199 which allows BRADEMAX (number of read errors tolerated by
200 dump) to be adjusted using the -I option.
202 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
203 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
206 6. Add the -l option to restore to specify if, when doing a
207 remote restore, the file used to access the backup is a
208 regular file (the defaults being a tape drive). Restore needs
209 to know this information when reading a remote compressed
210 dump. Previously, this information was autodetected, but
211 the autodetection code fails (with ioctl: Inappropriate ioctl
212 for device) when using a non Linux remote box. Thanks to
213 many users and especially to Eros Albertazzi
214 <eros@lamel.bo.cnr.it> for reporting this.
216 7. Found a workaroung for the dump deadlock problem (3 childs
217 stuck in pause(), father in read()). The workaround seems
218 to work for me and several beta-testers. If it doesn't work
219 for you, please report back.
221 8. Updated the RPM spec file (BuildPrereq, URL etc).
223 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
224 ========================================================================
226 1. Fixed the permissions of a newly created QFA file by dump.
228 2. Cleaned up the source of dump (the external variables
229 definition was a complete mess, making possible to have
232 3. Fixed restore to use the full tape volume path when doing
233 a compare (since it changes the working directory to the
234 filesystem being compared in the process).
236 4. Added the -q option to dump which makes dump abort
237 whenever it needs operator attention. It should be
238 easier to use dump in scripts with this option.
240 5. Detect the use of incompatible options to dump and
241 refuse them (like -a and -B options together).
243 6. Added bzip2 compression to dump/restore (use option -j level
244 to select it). Note that you will need a restore version
245 >= 0.4b24 in order to restore a bzip2 compressed dump.
246 The same warning as for the zlib compression applies:
247 the tape format of a bzip2 dump is not compatible with the
248 original BSD tape format.
250 7. Fixed a overflow problem in dump corrupting the dump when
251 very large files were encountered. Thanks to Vladimir Nadvornik
254 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
255 kernel buffer/page cache before starting the dump, helping
256 a bit those who use dump on mounted filesystems. Thanks to
257 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
260 9. Updated the RPM spec file following the RedHat changes
261 (dynamically linked binaries now in /usr/sbin etc).
263 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
264 which enables restore to recognise multi volume compressed dumps
265 done on CD. Included his scripts for dump (which pipe the dump
266 output directly into cdrecord) and restore. There is now
267 possible to to backups to CD on the fly!
269 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
270 ===================================================================
272 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
273 Marcus Meissner (Caldera International Security Dept.).
275 2. Implement the Sun rmt extended protocol. Patch submitted
276 by Ian Gordon <iangordon@users.sourceforge.net>.
278 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
279 the linux kernel header. This ensures that dump always has the
280 latest version of this file. Patch submitted by
281 Andreas Dilger <adilger@turbolinux.com>.
283 4. Report any filesystem present in either /etc/fstab with a
284 positive passno or /etc/dumpdates in dump -w output.
285 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
287 5. Fixed the looping problem in dump introduced in the
290 6. Changed the -B option of dump to limit the size of
291 _compressed_ output per volume if compression is on.
292 Patch contributed by Helmut Jarausch
293 <jarausch@igpm.rwth-aachen.de>. Note however that, since
294 it is impossible to predict the size of the compressed
295 data before doing the compression, the -B limit is a bit
298 7. Fixed a bug in reading the operator typed file/tape path for
299 the next volume in restore.
301 8. Implemented a "-F script" option to restore which permits the
302 user to specify a script which will be launched at the
303 beginning of each tape, useful for automatic programming of
304 tape changers for example. See the restore man page for the
305 script parameters and return codes.
307 9. Small fix for the QFA routines provided by Uwe Gohlke
308 <uwe@ugsoft.de>, and some recommendations for QFA uses in
311 10. Fixed the multivolume restoring where making a mistake
312 to the 'Mount next tape' prompt caused several blocks to
315 11. Enhanced the -e option of dump to take as a parameter a
316 comma separated list of inode numbers.
318 12. Added the -E option to dump which specify a file containing
319 inode numbers to exclude from the dump.
321 13. Fixed the compressed multi-volume dump + restore.
323 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
324 ==================================================================
326 1. Made dump report the number of blocks dumped per volume.
327 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
329 2. Fix a bug in dump where waiting too much at the 'change volumes'
330 question caused the volume to be rewritten. Thanks to
331 Graham Reed <greed@users.sourceforge.net> for reporting the
332 bug and providing a patch.
334 3. Added a compression option to dump, contributed by Jerry
335 Peters <gapeters@worldnet.att.net>.
337 WARNING: the compression works only on files/pipes and
338 drives supporting variable block size.
340 WARNING2: tapes created using the compression feature are
341 incompatible with the standard dump tape format, and a
342 version of restore >= 0.4b22 will be needed for extraction.
344 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
345 architectures. Thanks to Paul Slootman <paul@debian.org> for
346 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
349 5. Many cleanups (CPP defines, const char warnings, check of
350 ext2fs COMPAT flags, time_t cleanups, added libext2 version
351 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
353 6. Made --prefix option work in configure. All the install path
354 are now based on the configure parameters.
356 7. Added the Quick File Access mode in dump/restore, contributed
357 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
358 a file tape position for each inode, and this file is used by
359 restore (if called with parameter Q and the filename)
360 to directly position the tape at the file restore is currently
361 working on. This saves hours when restoring single files from
362 large backups, saves the tapes and the drive's head. Use
363 --enable-qfa option of configure to compile in the QFA support.
365 8. Added the possibility to dump several files and directories
366 in a single invocation of dump. Thanks to Uwe Gohlke
367 <uwe@ugsoft.de> for implementing this option.
369 9. Fixed the dumping and restoring of files with holes
370 (files ending with holes didn't get dumped or restored
373 10. Fixed a socket descriptor leak in dump, which leaved opened
374 3 file descriptors per dump process (and there is one dump
377 11. Fixed dump large file system support, by explicit use of
378 open64/lseek64/etc functions (explicit use needed because
379 e2fsprogs libraries don't behave well when compiled with
380 FILE_OFFSET_BITS=64).
382 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
383 ======================================================================
385 1. Fixed some bugs in the dump -w|-W logic introduced by
386 the previous version. Thanks to Andreas Dilger
387 <adilger@turbolinux.com> for his help on this one.
389 2. Fixed again a compilation problem when using old e2fs
390 headers (filesystem label related). Thanks to many users
391 who reported this stupid error.
393 3. Fixed a build problem on old lib5 based systems dealing with
394 _PATH_MNTTAB being undefined. Thanks to John Adams
395 <johna@onevista.com> for reporting the bug.
397 4. Improved the error detection in restore when creating
398 the temporary files in TMPDIR. Now restore will corectly
399 report a 'no space left on device' error instead of
400 strange errors which could imply an archive corruption.
401 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
402 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
404 5. Added the throughput information in dump's progression
405 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
408 6. Use libext2fs's inode scanning routines, which are particularly
409 robust when dealing with errors (especially when having some
410 bad blocks on the filesystem). Many thanks to Colin
411 <colin@horizon.com> for reporting the bug, investigating
412 the issues, finding a workaround, writing the patch and
413 fully testing it... (of course, if this patch will break
414 anything, he is to blame for :-)).
416 7. Made dump and restore LFS aware. Dump can dump a filesystem
417 containing large files, generate a large file on output and
418 restore can restore them. This feature is not enabled by
419 default, use --enable-largefile option of configure to enable
420 it (you need to have a LFS aware glibc though). Thanks to
421 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
422 and to Theodore T'so <tytso@valinux.com> for his always
425 8. Made dump ask upon a tape write error if it should rewrite
426 the current volume (assume this is a bad tape) or if it should
427 assume an end-of-tape condition (useful for tape drives which
428 behaves badly at the end of the tape). Thanks to Andreas
429 Dilger <adilger@turbolinux.com> for the suggestion.
431 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
432 =======================================================================
434 1. Fixed a small compilation problem due to a change
435 in the definintion of the struct sigaction in
436 glibc 2.0 / libc5. Thanks to Gunther Schlegel
437 <schlegel@riege.de> for reporting the bug and to
438 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
441 2. Modified the label and uuid analysis in order to be
442 self-contained (independant of kernel/libc headers). This
443 should fix the compile with older kernel/libc headers and
444 will preserve the functionality. Thanks to Bernhard Erdmann
445 <bernhard.erdmann@gmx.de> for reporting the bug.
447 3. The 'exclude inode' option, if applied to a directory
448 inode, excludes now correctly all the leaves of this
449 directory. Thanks to John R. Dennison
450 <gerdesas@users.sourceforge.net> for reporting the bug.
452 4. Fixed the '-e' option to disable the possibility
453 to exclude the root inode (which causes the dumps to
454 be unreadable by restore). Prevented array overflow
455 when multiple -e options are used.
457 5. Fixed dump to correctly interpret a filesystem argument
458 which is a mountpoint and it is not an absolute path
459 (as specified in the fstab). Thanks to Bernhard R. Erdmann
460 <be@berdmann.de> for reporting the bug.
462 6. Made dump able to backup files larger than 2 GB. Note that
463 dump still doesn't cope with files larger than 4 GB.
465 7. Restore the real uid and gid before invoking an external
466 RSH program (big hole when dump or restore is suid root!).
468 8. Get the values for the default tape device and the location
469 of fstab file from the system headers. Thanks to
470 Andreas Dilger <adilger@turbolinux.com> for the patch.
472 9. Made dump -w|-W report all recognized filesystems
473 present in either /etc/fstab or /etc/dumpdates, and present
474 the list in the same order as in fstab file. Thanks
475 to Andreas Dilger <adilger@turbolinux.com> for the patch.
477 10. Made dump's -a (automatic end of tape detection) the
478 default. Specifying one of -B, -c, -d or -s options will
479 override the EOM detection. Thanks to Andreas Dilger
480 <adilger@turbolinux.com> for the patch.
482 11. Save the ext2 filesystem label into the dump volume label.
483 Specifying a volume label on the command line (-L option)
484 will override this feature. Thanks to Andreas Dilger
485 <adilger@turbolinux.com> for the patch.
487 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
488 =====================================================================
490 1. Fixed the signal handling in dump (which I broke in 0.4b17)
491 which was causing several strange dump failures (dump
492 hanged or segmentation faults etc).
494 2. Specified the default blocksize in the dump man page.
496 3. Changed two info messages of restore to be written on stdout
497 instead of stderr in order to leave stderr only for errors
498 or warnings. Thanks to Stephen Atwell
499 <satwell@urbana.css.mot.com> for the suggestion.
501 4. Corrected an off by one calculation which prevented
502 dumping a directory having a 1 character name length.
503 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
504 for reporting the bug.
506 5. Reinforce dump in order to allow only level 0 and no
507 -u option when dumping a subdirectory, and document
508 this in the man page. Thanks to Bernhard Erdmann
509 <bernhard.erdmann@gmx.de> for reporting the bug.
511 6. Fixed a small harmless bug in restore which caused
512 in some conditions hard links to be restored several
513 times (therefore generation some warning messages).
514 Thanks to Kenneth Porter <shiva@well.com> for
517 7. Updated the RPM spec file to the latest RedHat version,
518 providing FHS packaging and other cosmetic changes.
519 You will need RPM version >= 3.0.5 in order to build the RPM.
521 8. Updated the configure script to check for libtermcap
522 before checking for libreadline, because we need this
523 library in order to compile the readline support.
525 9. Made dump understand the LABEL= and UUID= notation
526 both in /etc/fstab and on the command line. Note that
527 you will need the /proc filesystem in order to use
528 these notations. Thanks to Erik Troan <ewt@redhat.com>
529 for providing the patch.
531 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
532 ===================================================================
534 1. Fixed a potential buffer overflow in restore. Thanks
535 to Stan Bubrouski <satan@fastdial.net> for reporting
538 2. Fixed a readline-related bug which prevented
539 'cat DUMPFILE | restore -i -f -' from working. Thanks
540 to Charles Karney <karney@users.sourceforge.net>
543 3. Changed a few "panic" into "exit", causing restore to
544 be more stable against some attacks (like the last one
545 reported on Bugtraq, although the last version of restore
546 was not vulnerable - just dumped core). Thanks to
547 Andreas Hasenack <andreas@conectiva.com.br> for reporting
550 4. Removed the suid-root bit on dump and restore in the
551 default build (and generated RPMs). It should be safer
552 now. Users who need the suid-root capabilities in order
553 to make network backups should read first the man page
554 of dump and enable the suid bit manually.
556 5. Added -ltermcap to the compile parameters for restore
557 when using readline support, in order to make the compile
558 process work on some systems (SuSE ?). Thanks to
559 Patrik Schindler <poc@pocnet.net> for reporting the bug.
561 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
562 ====================================================================
564 1. The -F script is called now *only* at the end of a tape,
565 not at the end of the dump. Thanks to Christian Haul
566 <haul@informatik.tu-darmstadt.de> for the bug report.
568 Normally, the device name and the current volume number
569 are passed on the command line of the script. However,
570 if you want the old style script (with no arguments
571 passed on the command line), you can enable it in
572 configure with the --enable-oldstylefscript.
574 2. Use posix signal handling to preserve dump functionality
575 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
578 3. Made the exit code of restore in case of a 'C'ompare
579 command reflect the result of the compare. An exit status
580 of 0 means the dump archive is correct, 1 means tape errors,
581 2 means that some files were modified. Thanks to Kenneth Porter
582 <shiva@well.com> for the suggestion.
584 4. Made (finally) quotes work in interactive restore.
586 5. Small fixes in order to allow dump to compile with a
587 really old e2fsprogs version. Thanks to Ian Zimmerman
588 <itz@speakeasy.org> for the bug report.
590 6. Add GNU readline capabilities to interactive restore.
591 Use configure's --enable-readline flag to enable this feature.
592 Thanks to Patrik Schindler <poc@pocnet.net> for the
595 7. Do the compare on the fly in restore 'C' mode (this will
596 allow not to exhaust the available /tmp space when
597 ccmparing large files). Thanks to Kenneth Porter
598 <shiva@well.com> for the suggestion.
600 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
601 ====================================================================
603 1. Fixed some several duplicate 'DUMP: DUMP:' in the
606 2. Corrected the estimation of blocks to dump. Note that
607 this is still wrong for large directory inodes, due
608 to the size of a BSD directory entry on the tape
609 (estimation is based on the size of the ext2 directory,
610 which is smaller because it doesn't contain the null
611 character at the end).
613 3. Made dump report the total number of MB written to
614 tape at the end of run. Thanks to W. Reilly Cooley
615 <wcooley@nakedape.cc> for the patch.
617 4. Added the -X option to restore giving the possibility
618 to read the names of the files to be extracted/listed
619 from a text file (in addition of the command line).
620 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
623 5. Added the device name and the current volume number
624 as arguments to the end of tape script (dump -F option).
626 6. Made the multi-volume dump work again (after having
627 broken it in 0.4b15).
629 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
630 ===================================================================
632 1. Added a prompt command in interactive restore mode. Thanks
633 to Andreas Dilger <adilger@home.com> for the patch.
635 2. Fixed a buffer overflow problem in dump (caused by
636 not checking the size of the filesystem parameter).
637 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
638 reporting this on Bugtraq (and to several dump users
639 who forwarded me his mail).
641 3. Added the '-F script' option to dump in order to
642 launch a script at the end of each tape (to be used
643 with a tape changer, or to notify the sysadmin by
646 4. Fixed a bug in restore compare code caused by the changes
649 5. Fixed the treatment of options using the old BSD syntax
650 in both dump and restore.
652 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
653 =======================================================================
655 1. Fixed a bug in dump which may have caused invalid deleted
656 directories to be dumped out if they were deleted after the
657 mapping pass. This could occure on active filesystem and lead
658 to heap corruption (causing dump malfunction in many possible ways).
659 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
660 bug and submitting the patch.
662 2. Fixed the handling of the filesystem parameter in dump. A
663 '/mnt/disk' parameter caused the disk contents to be dumped,
664 but a '/mnt/disk/' parameter caused the mountpoint directory
665 to be dumped (generally an empty directory).
667 3. Improved the output of dump in order to tell which directory
668 it is currently dumping (when dumping a subtree).
670 4. Added the '-e' exclude inode option to dump. Thanks to
671 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
673 5. Added a REPORTING-BUGS file in order to provide a guide
674 on how to correctly report a bug in dump/restore.
676 6. Fixed a restore problem when restoring a hard link to an inode
677 having the immutable or append-only attribute set. Thanks to
678 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
680 7. Fixed a compatibility problem between dumps made on little
681 endian machines (the format was correct) and big endian
682 machines (the format was incorrect). This fix break the
683 compatibility with the older dumps made on big endian
684 machines (sparc, m86k, ppc etc). For the first time in
685 linux dump's history (I believe), the dumps made by *BSD,
686 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
687 so interchangeable. Thanks to Rob Cermak
688 <cermak@ahab.rutgers.edu> for submitting the bug and
689 helping me test the fix.
691 8. Fixed the way dump reports the remaining percent/time, if
692 the number of blocks actually dumped exceeds the estimated
693 number of blocks. Thanks to Jean-Paul van der Jagt
694 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
696 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
697 ======================================================================
699 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
700 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
702 2. Removed the requirement to build the RPM as root from the
703 spec file. Thanks to Christian Weisgerber
704 <naddy@mips.rhein-neckar.de> for submitting this.
706 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
707 causing dump to block when dumping really huge filesystems.
708 Many thanks to Patrik Schindler <poc@pocnet.net> for
709 helping me find this bug.
711 4. Fixed the treatment for an interrupt signal when dump access
712 the remote tape through RSH. Thanks to Christian Weisgerber
713 <naddy@mips.rhein-neckar.de> for providing the patch.
715 5. Fixed a bug which was causing dump/restore to display
716 garbage characters instead of the remote host name.
718 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
719 =====================================================================
721 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
722 <kukuk@suse.de> for submitting the patch.
724 2. Fix for the exit code when using the size estimate option of
725 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
726 submitting the patch.
728 3. Handle EINTR in atomical reads/writes in dump, which was causing
729 dump to fail on some systems. Thanks to Eric Jergensen
730 <eric@dvns.com> for reporting the bug and submitting the patch.
732 4. Handle more than 16 characters for the device names in dumpdates.
733 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
734 tracking down the problem and proposing the solution.
736 5. Fixed a bug in dump which prevented the creation of the
737 dumpdates file when doing a 0-level dump without already
738 having a dumpdates file. Thanks to Patrik Schindler
739 <poc@pocnet.net> for reporting the bug.
741 6. Changed the way dump 'S' flag reports the size estimate
742 from number of blocks into bytes (making it compatible
743 with the Solaris version, and simplifying things for
744 amanda users). Thanks to Jason L Tibbitts III
745 <tibbs@math.uh.edu> for reporting the bug.
747 7. Fixed a compatibility problem in linux/alpha dump tape format.
748 Now the linux/alpha dump are (again) compatible with the
749 other dump formats. But this breaks compatibility with
750 older dumps made on alpha. Thanks to Mike Tibor
751 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
753 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
754 ======================================================================
756 1. Added a '--enable-kerberos' to configure.
758 2. Added a 'S' option to dump which determines the amount of space
759 that is needed to perform the dump without actually doing it, similar
760 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
761 <cermak@ahab.rutgers.edu>.
763 3. Added a 'M' multi-volume option to dump and restore which enables
764 dumping to multiple files (useful when dumping to an ext2
765 partition to make several dump files in order to bypass the 2GB
766 file size limitation). The 'f' argument is treated as a prefix and
767 the output files will be named <prefix>001, <prefix>002 etc. With
768 the 'M' flag, restore automatically selects the right file without
769 asking to enter a new tape each time.
771 4. Fixed a memory leak which was causing dump to grow very big
772 (270MB when dumping a 10GB filesystem...). Thanks to Jason
773 Fearon <jasonf@netrider.org.au> for reporting the bug.
775 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
776 ======================================================================
778 1. Make configure test if the system glob routines support
779 extended syntax (ALTDIRFUNC). If not, use the internal glob
780 routines instead of system ones. Thanks to Bernhard Sadlowski
781 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
782 and helping me resolve this and other minor libc5 compiling
785 2. Fix a problem when dumping a ext2fs with the 'filetype'
786 feature enabled. Thanks to Patrick J. LoPresti
787 <patl@cag.lcs.mit.edu> for reporting the bug and to
788 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
790 3. Made the nodump flag work on directories. A directory which
791 has the nodump flag gets never dumped, regardless of its
794 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
795 which allows dump on an active ext3 filesystem. However, this
796 is a "quick and dirty" patch which enables backup of an ext3
797 filesystem through the ext2 compatibility (by ignoring the
798 NEEDS_RECOVERY bit). The journal file is not recognized and
799 it is dumped (it should not).
801 5. Test the superblock compatibility flags when dumping, in order
802 to be sure that we know how to deal with specific features.
804 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
805 ====================================================================
807 1. Use lchown instead of chown, fixing a possible security problem
808 when restoring symlinks (a malicious user could use this
809 to deliberately corrupt the ownership of important system files).
810 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
811 this and providing the patch.
813 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
814 ====================================================================
816 1. Put dump sources under CVS, added Id tags in all files so
817 one can use 'ident' on binary files.
819 2. Added the dump/restore version in the usage text so one can
820 easily verify the version he is using.
822 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
823 a va_start/va_end problem on linux-ppc (always call va_start
824 va_end in pairs each time we use a vararg function).
826 4. Added again the DT_* constants because old libc does not
827 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
828 for submitting the bug report.
830 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
831 this should enable dumping big (huge) filesystems.
833 6. Added the RSH environment variable in order to be able to
834 use a rsh replacement like ssh when doing remote backups (and
835 bypass the security limitations of rcmd). Now you can do remote
836 backups without being root (or making dump setuid root).
838 7. Modified again the way dumpdates works. For incremental dumps,
839 we need to read dumpdates even if we are not using 'u' option.
840 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
843 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
844 ===================================================================
846 1. Removed the 'k' flag from the restore 'about' text if kerberos
849 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
850 to them (fsetflags takes a char*, setflags an open fd!).
852 3. (f)setflags is called only if the flags aren't empty. If the
853 file is a special file, a warning is printed, because changing
854 flags implies opening the device. Normally, a special file
855 should not have any flag... (Debian bug #29775, patch provided
856 by Abhijit Dasgupta <abhijit@ans.net>).
858 4. Made possible to dump a file system not mentioned in /etc/fstab.
859 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
861 5. Changed the default behaviour to not create dumpdates
862 unless 'u' option is specified. Removed the old "debian-patch"
863 which provided the same thing. (Debian bug #38136, #33818).
865 6. Removed all those dump*announce, since they were getting old...
867 7. Added warning messages if dumpdates does not exist and
868 when an inferior level dump does not exist (except for a level 0
871 8. Debugged the glob calls in interactive mode: restore used a
872 dirent struct which was different from the /usr/include/dirent.h
873 one (this used to work, is it a glibc2 change?), so none of the
874 compat glob (which used /usr/include/dirent.h) or the system glob
875 worked. Restore use now the system dirent (and the system
876 DT_* constants), which are compatible with BSD ones.
878 9. Added a configure flag (--with-dumpdatespath) to specify
879 the location of dumpdates file. By default, it is
882 10. Added the "AUTHOR" and "AVAILABILITY" sections and
883 included the current date/version in man pages.
885 11. Corrected the estimation of remaining time when
886 the operator doesn't change the tapes quickly enough. This
887 was an old bug, I thought I corrected it, and discovered
888 that in fact it was corrected in two different places, so
889 the results canceled each other...
891 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
892 ===================================================================
894 1. Integrated multiple patches from RedHat, Debian and SuSE:
896 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
898 - several fixes in the man pages.
899 - update the default tape device to /dev/st0.
900 - many updates for Linux Alpha (byte ordering, size_t etc).
902 - use environment variable for TMPDIR (instead of /tmp).
903 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
904 - workaround egcs bug (RedHat bugs #4281 and #2989).
905 - wire $(OPT) throughout Makefile's.
907 2. Upgrade the dump revision to 1, making possible to dump filesystems
908 made with e2fsprogs-1.15 or newer. Nothing seems to break...
910 3. Fix some compile warnings, prototype all functions.
912 4. Use glibc err/glob instead of internal compatibility
913 routines (only if available).
915 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
916 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
918 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
919 ======================================================================
921 1. Integrated the changes from FreeBSD-3.1-RELEASE
922 (mostly bug fixes, buffer overruns, dump has now an "automatic
923 tape length calculation" flag, dump/restore can use kerberos now
924 (this is NOT tested), use environment variables for TAPE and
927 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
929 3. Corrected a bug in restore when using 'C' option with multi-volumes
930 tapes (files splited accros two tapes give "size changed" errors
933 4. Corrected the long standing bug when dumping multiple tapes.
934 This works for me, needs further testing.
936 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
937 ====================================================================
939 1. Dump now runs correctly on kernels 2.1.x
940 Fix made by Gerald Peters <gapeters@worldnet.att.net>
942 Changes between versions 0.4b2 and 0.4b3
943 ========================================
945 1. Use realpath() if available
949 Changes between versions 0.4b1 and 0.4b2
950 ========================================
952 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
955 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
957 3. Fixed configure.in to correctly handle --enable-rmt
959 Changes between versions 0.3 and 0.4b1
960 ======================================
962 1. Integrated the changes from 4.4BSD-Lite2
964 2. Integrated the patches from Debian and Red Hat
966 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
968 4. Changed dump to use the Ext2fs library to get block addresses. This
969 should solve the endianness problem on SparcLinux.
971 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
972 distribution's one) to use autoconf
974 6. Fixed a few minor bugs
976 Changes between versions 0.2e and 0.2f
977 ======================================
979 1. Added the creation of named pipes (fifos) in restore.
981 2. Added the -N flag in the restore manual page.
983 3. Added the file kernel.patch which contains the llseek() optimization
984 patch for 1.2.x kernels.
986 4. Fixed a bug in the restoration of symbolic links: owner and group were
989 5. Integrated some changes from FreeBSD 2.2.
991 6. Added a call to ftruncate() after restoring each file to restore
992 correctly files ending by a hole.
994 Changes between versions 0.2d and 0.2e
995 ======================================
997 1. Fixed a bug in the "set owner/mode" process. Every file was restored
998 with owner = root (0) and group = root/wheel/whatever (0).
1000 Changes between versions 0.2c and 0.2d
1001 ======================================
1003 1. Dump is now able to backup 2GB+ filesystems.
1005 2. Dump and restore can now be linked as static binaries.
1007 Changes between versions 0.2b and 0.2c
1008 ======================================
1010 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1012 Changes between versions 0.2a and 0.2b
1013 ======================================
1015 1. Really fixed the bug that I should have corrected in 0.2a.
1017 2. Enabled optimization again.
1019 Changes between versions 0.2 and 0.2a
1020 =====================================
1022 1. Disabled the optimization during compilation.
1024 Changes between versions 0.1 and 0.2
1025 ====================================
1027 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1028 the fs_type field (actually, I modified the file fstab.c to make it
1029 use the mntent functions).
1031 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1032 made by dump should be readable by the BSD restore and backups made
1033 by the BSD dump should be readable by restore. Unfortunately, this
1034 means that the dump format has changed between version 0.1 and version
1037 3. Dump is now able to backup a subtree, it is no longer limited to whole
1038 filesystems like the BSD version.
1040 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1043 Changes between versions 0.0 and 0.1
1044 ====================================
1046 1. Now create links rdump and rrestore during the `make install' step.
1048 2. Linux port specific bugs added to the manual pages
1050 3. Incorrect estimation of the number of tapes blocks fixed when doing
1051 incremental backups.
1053 4. Better ls-like format in restore in interactive mode.