1 $Id: CHANGES,v 1.175 2002/05/16 21:22:36 stelian Exp $
3 Changes between versions 0.4b28 and 0.4b29 (released ??????????????)
4 ====================================================================
6 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
7 non Linux clients were misinterpreted. The description of the problem
8 (incompatible numbering in Linux mtio opcodes) is documented at
9 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
10 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
11 providing an excellent, cross-platform replacement for rmt in his
14 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
15 which caused restore to fail when the inode maps were not entirely
16 contained on the first volume (when using really small volumes or
17 when dumping a huge number of inodes).
19 3. Fixed a problem in dump, where files in subdirectories of directories
20 that have the nodump flag set (or being excluded with -e/-E) were
21 sometimes incorrectly dumped (instead of being excluded). The original
22 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
23 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
24 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
25 help testing the patch.
27 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
28 ====================================================================
30 1. Fixed a bug in the QFA feature which could have caused
31 a bad tape positionning on restore (causing extra delays in
32 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
34 2. Added a small note in the dump man page specifying that
35 there should be no white space between the option letter and
36 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
38 3. Made restore work with older versions of the readline library,
39 by checking in configure for several extended symbols. Restore
40 can now be compiled with a readline as old as the 2.0 release
41 (though it may be a good idea to upgrade it to a more recent
42 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
43 reporting the build failures.
45 4. Fixed a performance problem with the QFA file creation in
46 dump, which made unnecessary seeks on the tape slowing down
47 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
50 5. Removed the inclusion of some kernel headers in the dump
51 source, which prevented the compile in some kernel/glibc
52 headers/architecture combination. Thanks to Bdale Garbee
53 <bdale@gag.com> for reporting the bug.
55 6. Added the appropriate error message when dump fails to
56 open the output file for writing. Thanks to Amith Varghese
57 <amithv@yahoo.com> for reporting this bug.
59 7. Made restore able to understand large Solaris ufsdump tapes
60 (containing inodes bigger than 4194304). Sun have introduced
61 an "extension" to the dump tape format when dealing with
62 those inodes, which was uncorrectly handled by Linux restore.
63 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
64 providing a test case.
66 8. Added the -m parameter to dump which optimises the output for
67 inodes having been changed but not modified since the last dump
68 ('changed' and 'modified' have the meaning defined in stat(2)).
69 For those inodes, dump will save only the metadata, instead of
70 saving the entire inode contents. Inodes which are either
71 directories or have been modified since the last dump are saved
72 in a regular way. Uses of this flag must be consistent, meaning
73 that either every dump in an incremental dump set have the flag,
76 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
77 =======================================================================
79 1. Fixed behaviour of dump when exceeding resource limits
82 2. Added the -L flag to restore to allow the user to specify a
83 maximal allowed number of miscompares when using restore
84 with the -C option to check the backup.
86 3. Detailed the manual entry for the -N option of restore.
88 4. Added the -a flag to restore to make able doing unattended
89 restores in -i or -x mode (automatically walks through the
90 multiple dump volumes).
92 5. Extended the QFA mode to work with local files and/or
93 remote tapes and files. This way, restore can know in advance
94 the tape number and the offset for the inodes to extract and
95 can minimize the extraction time by seeking directly to the
98 6. Added the -A <archive> option to both dump and restore,
99 which makes dump to archive a dump table-of-contents in
100 the specified file to be used by restore to determine
101 whether a file is in the dump file that is being restored.
102 (the archive file syntax is also compatible with the
103 Solaris ufsdump generated one).
105 7. Small fix in restore making it able to read some (broken ?)
106 Solaris ufsdump tapes.
108 8. Fixed dump to correctly recognise the root filesystem when using
109 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
110 for reporting this bug.
112 9. Added the -P <file> option to restore to create a
113 Quick File Access file from an already made dump. Patch
114 contributed by Uwe Gohlke <uwe@ugsoft.de>.
116 10. Made restore compile and run on Solaris, making it a
117 possible replacement for the standard ufsrestore. Port was
118 contributed by Uwe Gohlke <uwe@ugsoft.de>.
120 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
121 =====================================================================
123 1. Added a set of backup scripts from Eugenio Diaz
124 <getnito@yahoo.com> in the examples section. It features
125 automatic (cron based) full and incremental dumping of
126 several filesystems on a separate filesystem.
128 2. Fixed a off-by-one miscalculation which disabled dumping
129 a one letter subdirectory. Thanks to Chris Siebenmann
130 <cks@utcc.utoronto.ca> for reporting the bug.
132 3. Fixed several restore bugs occuring when trying to
133 restore some missing files on the tape. Thanks to Chris
134 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
136 4. Fixed --with-ldopts configure argument passing, installing from
137 a separate object directory, makefile cleanups contributed
138 by <splite@purdue.edu>.
140 5. Fix a bug which could caused, in some conditions, the highest
141 number inode of a filesystem, to not be dumped. Many thanks
142 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
145 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
146 =======================================================================
148 1. Added a mini howto from Patrick Walsh in the examples
151 2. Minor man pages syntax corrections. Thanks to
152 Chris Danis <screechco@home.com> for reporting the bugs.
154 3. Added a script from David B. Peterson <dave@toppledwagon.com>
155 to the examples section. It features dumping several
156 filesystems to a remote tape drive upon ssh.
158 4. Added a patch provided by Richard Jones <rich@annexia.org>
159 which allows BRADEMAX (number of read errors tolerated by
160 dump) to be adjusted using the -I option.
162 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
163 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
166 6. Add the -l option to restore to specify if, when doing a
167 remote restore, the file used to access the backup is a
168 regular file (the defaults being a tape drive). Restore needs
169 to know this information when reading a remote compressed
170 dump. Previously, this information was autodetected, but
171 the autodetection code fails (with ioctl: Inappropriate ioctl
172 for device) when using a non Linux remote box. Thanks to
173 many users and especially to Eros Albertazzi
174 <eros@lamel.bo.cnr.it> for reporting this.
176 7. Found a workaroung for the dump deadlock problem (3 childs
177 stuck in pause(), father in read()). The workaround seems
178 to work for me and several beta-testers. If it doesn't work
179 for you, please report back.
181 8. Updated the RPM spec file (BuildPrereq, URL etc).
183 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
184 ========================================================================
186 1. Fixed the permissions of a newly created QFA file by dump.
188 2. Cleaned up the source of dump (the external variables
189 definition was a complete mess, making possible to have
192 3. Fixed restore to use the full tape volume path when doing
193 a compare (since it changes the working directory to the
194 filesystem being compared in the process).
196 4. Added the -q option to dump which makes dump abort
197 whenever it needs operator attention. It should be
198 easier to use dump in scripts with this option.
200 5. Detect the use of incompatible options to dump and
201 refuse them (like -a and -B options together).
203 6. Added bzip2 compression to dump/restore (use option -j level
204 to select it). Note that you will need a restore version
205 >= 0.4b24 in order to restore a bzip2 compressed dump.
206 The same warning as for the zlib compression applies:
207 the tape format of a bzip2 dump is not compatible with the
208 original BSD tape format.
210 7. Fixed a overflow problem in dump corrupting the dump when
211 very large files were encountered. Thanks to Vladimir Nadvornik
214 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
215 kernel buffer/page cache before starting the dump, helping
216 a bit those who use dump on mounted filesystems. Thanks to
217 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
220 9. Updated the RPM spec file following the RedHat changes
221 (dynamically linked binaries now in /usr/sbin etc).
223 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
224 which enables restore to recognise multi volume compressed dumps
225 done on CD. Included his scripts for dump (which pipe the dump
226 output directly into cdrecord) and restore. There is now
227 possible to to backups to CD on the fly!
229 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
230 ===================================================================
232 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
233 Marcus Meissner (Caldera International Security Dept.).
235 2. Implement the Sun rmt extended protocol. Patch submitted
236 by Ian Gordon <iangordon@users.sourceforge.net>.
238 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
239 the linux kernel header. This ensures that dump always has the
240 latest version of this file. Patch submitted by
241 Andreas Dilger <adilger@turbolinux.com>.
243 4. Report any filesystem present in either /etc/fstab with a
244 positive passno or /etc/dumpdates in dump -w output.
245 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
247 5. Fixed the looping problem in dump introduced in the
250 6. Changed the -B option of dump to limit the size of
251 _compressed_ output per volume if compression is on.
252 Patch contributed by Helmut Jarausch
253 <jarausch@igpm.rwth-aachen.de>. Note however that, since
254 it is impossible to predict the size of the compressed
255 data before doing the compression, the -B limit is a bit
258 7. Fixed a bug in reading the operator typed file/tape path for
259 the next volume in restore.
261 8. Implemented a "-F script" option to restore which permits the
262 user to specify a script which will be launched at the
263 beginning of each tape, useful for automatic programming of
264 tape changers for example. See the restore man page for the
265 script parameters and return codes.
267 9. Small fix for the QFA routines provided by Uwe Gohlke
268 <uwe@ugsoft.de>, and some recommendations for QFA uses in
271 10. Fixed the multivolume restoring where making a mistake
272 to the 'Mount next tape' prompt caused several blocks to
275 11. Enhanced the -e option of dump to take as a parameter a
276 comma separated list of inode numbers.
278 12. Added the -E option to dump which specify a file containing
279 inode numbers to exclude from the dump.
281 13. Fixed the compressed multi-volume dump + restore.
283 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
284 ==================================================================
286 1. Made dump report the number of blocks dumped per volume.
287 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
289 2. Fix a bug in dump where waiting too much at the 'change volumes'
290 question caused the volume to be rewritten. Thanks to
291 Graham Reed <greed@users.sourceforge.net> for reporting the
292 bug and providing a patch.
294 3. Added a compression option to dump, contributed by Jerry
295 Peters <gapeters@worldnet.att.net>.
297 WARNING: the compression works only on files/pipes and
298 drives supporting variable block size.
300 WARNING2: tapes created using the compression feature are
301 incompatible with the standard dump tape format, and a
302 version of restore >= 0.4b22 will be needed for extraction.
304 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
305 architectures. Thanks to Paul Slootman <paul@debian.org> for
306 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
309 5. Many cleanups (CPP defines, const char warnings, check of
310 ext2fs COMPAT flags, time_t cleanups, added libext2 version
311 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
313 6. Made --prefix option work in configure. All the install path
314 are now based on the configure parameters.
316 7. Added the Quick File Access mode in dump/restore, contributed
317 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
318 a file tape position for each inode, and this file is used by
319 restore (if called with parameter Q and the filename)
320 to directly position the tape at the file restore is currently
321 working on. This saves hours when restoring single files from
322 large backups, saves the tapes and the drive's head. Use
323 --enable-qfa option of configure to compile in the QFA support.
325 8. Added the possibility to dump several files and directories
326 in a single invocation of dump. Thanks to Uwe Gohlke
327 <uwe@ugsoft.de> for implementing this option.
329 9. Fixed the dumping and restoring of files with holes
330 (files ending with holes didn't get dumped or restored
333 10. Fixed a socket descriptor leak in dump, which leaved opened
334 3 file descriptors per dump process (and there is one dump
337 11. Fixed dump large file system support, by explicit use of
338 open64/lseek64/etc functions (explicit use needed because
339 e2fsprogs libraries don't behave well when compiled with
340 FILE_OFFSET_BITS=64).
342 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
343 ======================================================================
345 1. Fixed some bugs in the dump -w|-W logic introduced by
346 the previous version. Thanks to Andreas Dilger
347 <adilger@turbolinux.com> for his help on this one.
349 2. Fixed again a compilation problem when using old e2fs
350 headers (filesystem label related). Thanks to many users
351 who reported this stupid error.
353 3. Fixed a build problem on old lib5 based systems dealing with
354 _PATH_MNTTAB being undefined. Thanks to John Adams
355 <johna@onevista.com> for reporting the bug.
357 4. Improved the error detection in restore when creating
358 the temporary files in TMPDIR. Now restore will corectly
359 report a 'no space left on device' error instead of
360 strange errors which could imply an archive corruption.
361 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
362 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
364 5. Added the throughput information in dump's progression
365 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
368 6. Use libext2fs's inode scanning routines, which are particularly
369 robust when dealing with errors (especially when having some
370 bad blocks on the filesystem). Many thanks to Colin
371 <colin@horizon.com> for reporting the bug, investigating
372 the issues, finding a workaround, writing the patch and
373 fully testing it... (of course, if this patch will break
374 anything, he is to blame for :-)).
376 7. Made dump and restore LFS aware. Dump can dump a filesystem
377 containing large files, generate a large file on output and
378 restore can restore them. This feature is not enabled by
379 default, use --enable-largefile option of configure to enable
380 it (you need to have a LFS aware glibc though). Thanks to
381 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
382 and to Theodore T'so <tytso@valinux.com> for his always
385 8. Made dump ask upon a tape write error if it should rewrite
386 the current volume (assume this is a bad tape) or if it should
387 assume an end-of-tape condition (useful for tape drives which
388 behaves badly at the end of the tape). Thanks to Andreas
389 Dilger <adilger@turbolinux.com> for the suggestion.
391 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
392 =======================================================================
394 1. Fixed a small compilation problem due to a change
395 in the definintion of the struct sigaction in
396 glibc 2.0 / libc5. Thanks to Gunther Schlegel
397 <schlegel@riege.de> for reporting the bug and to
398 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
401 2. Modified the label and uuid analysis in order to be
402 self-contained (independant of kernel/libc headers). This
403 should fix the compile with older kernel/libc headers and
404 will preserve the functionality. Thanks to Bernhard Erdmann
405 <bernhard.erdmann@gmx.de> for reporting the bug.
407 3. The 'exclude inode' option, if applied to a directory
408 inode, excludes now correctly all the leaves of this
409 directory. Thanks to John R. Dennison
410 <gerdesas@users.sourceforge.net> for reporting the bug.
412 4. Fixed the '-e' option to disable the possibility
413 to exclude the root inode (which causes the dumps to
414 be unreadable by restore). Prevented array overflow
415 when multiple -e options are used.
417 5. Fixed dump to correctly interpret a filesystem argument
418 which is a mountpoint and it is not an absolute path
419 (as specified in the fstab). Thanks to Bernhard R. Erdmann
420 <be@berdmann.de> for reporting the bug.
422 6. Made dump able to backup files larger than 2 GB. Note that
423 dump still doesn't cope with files larger than 4 GB.
425 7. Restore the real uid and gid before invoking an external
426 RSH program (big hole when dump or restore is suid root!).
428 8. Get the values for the default tape device and the location
429 of fstab file from the system headers. Thanks to
430 Andreas Dilger <adilger@turbolinux.com> for the patch.
432 9. Made dump -w|-W report all recognized filesystems
433 present in either /etc/fstab or /etc/dumpdates, and present
434 the list in the same order as in fstab file. Thanks
435 to Andreas Dilger <adilger@turbolinux.com> for the patch.
437 10. Made dump's -a (automatic end of tape detection) the
438 default. Specifying one of -B, -c, -d or -s options will
439 override the EOM detection. Thanks to Andreas Dilger
440 <adilger@turbolinux.com> for the patch.
442 11. Save the ext2 filesystem label into the dump volume label.
443 Specifying a volume label on the command line (-L option)
444 will override this feature. Thanks to Andreas Dilger
445 <adilger@turbolinux.com> for the patch.
447 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
448 =====================================================================
450 1. Fixed the signal handling in dump (which I broke in 0.4b17)
451 which was causing several strange dump failures (dump
452 hanged or segmentation faults etc).
454 2. Specified the default blocksize in the dump man page.
456 3. Changed two info messages of restore to be written on stdout
457 instead of stderr in order to leave stderr only for errors
458 or warnings. Thanks to Stephen Atwell
459 <satwell@urbana.css.mot.com> for the suggestion.
461 4. Corrected an off by one calculation which prevented
462 dumping a directory having a 1 character name length.
463 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
464 for reporting the bug.
466 5. Reinforce dump in order to allow only level 0 and no
467 -u option when dumping a subdirectory, and document
468 this in the man page. Thanks to Bernhard Erdmann
469 <bernhard.erdmann@gmx.de> for reporting the bug.
471 6. Fixed a small harmless bug in restore which caused
472 in some conditions hard links to be restored several
473 times (therefore generation some warning messages).
474 Thanks to Kenneth Porter <shiva@well.com> for
477 7. Updated the RPM spec file to the latest RedHat version,
478 providing FHS packaging and other cosmetic changes.
479 You will need RPM version >= 3.0.5 in order to build the RPM.
481 8. Updated the configure script to check for libtermcap
482 before checking for libreadline, because we need this
483 library in order to compile the readline support.
485 9. Made dump understand the LABEL= and UUID= notation
486 both in /etc/fstab and on the command line. Note that
487 you will need the /proc filesystem in order to use
488 these notations. Thanks to Erik Troan <ewt@redhat.com>
489 for providing the patch.
491 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
492 ===================================================================
494 1. Fixed a potential buffer overflow in restore. Thanks
495 to Stan Bubrouski <satan@fastdial.net> for reporting
498 2. Fixed a readline-related bug which prevented
499 'cat DUMPFILE | restore -i -f -' from working. Thanks
500 to Charles Karney <karney@users.sourceforge.net>
503 3. Changed a few "panic" into "exit", causing restore to
504 be more stable against some attacks (like the last one
505 reported on Bugtraq, although the last version of restore
506 was not vulnerable - just dumped core). Thanks to
507 Andreas Hasenack <andreas@conectiva.com.br> for reporting
510 4. Removed the suid-root bit on dump and restore in the
511 default build (and generated RPMs). It should be safer
512 now. Users who need the suid-root capabilities in order
513 to make network backups should read first the man page
514 of dump and enable the suid bit manually.
516 5. Added -ltermcap to the compile parameters for restore
517 when using readline support, in order to make the compile
518 process work on some systems (SuSE ?). Thanks to
519 Patrik Schindler <poc@pocnet.net> for reporting the bug.
521 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
522 ====================================================================
524 1. The -F script is called now *only* at the end of a tape,
525 not at the end of the dump. Thanks to Christian Haul
526 <haul@informatik.tu-darmstadt.de> for the bug report.
528 Normally, the device name and the current volume number
529 are passed on the command line of the script. However,
530 if you want the old style script (with no arguments
531 passed on the command line), you can enable it in
532 configure with the --enable-oldstylefscript.
534 2. Use posix signal handling to preserve dump functionality
535 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
538 3. Made the exit code of restore in case of a 'C'ompare
539 command reflect the result of the compare. An exit status
540 of 0 means the dump archive is correct, 1 means tape errors,
541 2 means that some files were modified. Thanks to Kenneth Porter
542 <shiva@well.com> for the suggestion.
544 4. Made (finally) quotes work in interactive restore.
546 5. Small fixes in order to allow dump to compile with a
547 really old e2fsprogs version. Thanks to Ian Zimmerman
548 <itz@speakeasy.org> for the bug report.
550 6. Add GNU readline capabilities to interactive restore.
551 Use configure's --enable-readline flag to enable this feature.
552 Thanks to Patrik Schindler <poc@pocnet.net> for the
555 7. Do the compare on the fly in restore 'C' mode (this will
556 allow not to exhaust the available /tmp space when
557 ccmparing large files). Thanks to Kenneth Porter
558 <shiva@well.com> for the suggestion.
560 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
561 ====================================================================
563 1. Fixed some several duplicate 'DUMP: DUMP:' in the
566 2. Corrected the estimation of blocks to dump. Note that
567 this is still wrong for large directory inodes, due
568 to the size of a BSD directory entry on the tape
569 (estimation is based on the size of the ext2 directory,
570 which is smaller because it doesn't contain the null
571 character at the end).
573 3. Made dump report the total number of MB written to
574 tape at the end of run. Thanks to W. Reilly Cooley
575 <wcooley@nakedape.cc> for the patch.
577 4. Added the -X option to restore giving the possibility
578 to read the names of the files to be extracted/listed
579 from a text file (in addition of the command line).
580 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
583 5. Added the device name and the current volume number
584 as arguments to the end of tape script (dump -F option).
586 6. Made the multi-volume dump work again (after having
587 broken it in 0.4b15).
589 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
590 ===================================================================
592 1. Added a prompt command in interactive restore mode. Thanks
593 to Andreas Dilger <adilger@home.com> for the patch.
595 2. Fixed a buffer overflow problem in dump (caused by
596 not checking the size of the filesystem parameter).
597 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
598 reporting this on Bugtraq (and to several dump users
599 who forwarded me his mail).
601 3. Added the '-F script' option to dump in order to
602 launch a script at the end of each tape (to be used
603 with a tape changer, or to notify the sysadmin by
606 4. Fixed a bug in restore compare code caused by the changes
609 5. Fixed the treatment of options using the old BSD syntax
610 in both dump and restore.
612 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
613 =======================================================================
615 1. Fixed a bug in dump which may have caused invalid deleted
616 directories to be dumped out if they were deleted after the
617 mapping pass. This could occure on active filesystem and lead
618 to heap corruption (causing dump malfunction in many possible ways).
619 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
620 bug and submitting the patch.
622 2. Fixed the handling of the filesystem parameter in dump. A
623 '/mnt/disk' parameter caused the disk contents to be dumped,
624 but a '/mnt/disk/' parameter caused the mountpoint directory
625 to be dumped (generally an empty directory).
627 3. Improved the output of dump in order to tell which directory
628 it is currently dumping (when dumping a subtree).
630 4. Added the '-e' exclude inode option to dump. Thanks to
631 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
633 5. Added a REPORTING-BUGS file in order to provide a guide
634 on how to correctly report a bug in dump/restore.
636 6. Fixed a restore problem when restoring a hard link to an inode
637 having the immutable or append-only attribute set. Thanks to
638 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
640 7. Fixed a compatibility problem between dumps made on little
641 endian machines (the format was correct) and big endian
642 machines (the format was incorrect). This fix break the
643 compatibility with the older dumps made on big endian
644 machines (sparc, m86k, ppc etc). For the first time in
645 linux dump's history (I believe), the dumps made by *BSD,
646 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
647 so interchangeable. Thanks to Rob Cermak
648 <cermak@ahab.rutgers.edu> for submitting the bug and
649 helping me test the fix.
651 8. Fixed the way dump reports the remaining percent/time, if
652 the number of blocks actually dumped exceeds the estimated
653 number of blocks. Thanks to Jean-Paul van der Jagt
654 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
656 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
657 ======================================================================
659 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
660 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
662 2. Removed the requirement to build the RPM as root from the
663 spec file. Thanks to Christian Weisgerber
664 <naddy@mips.rhein-neckar.de> for submitting this.
666 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
667 causing dump to block when dumping really huge filesystems.
668 Many thanks to Patrik Schindler <poc@pocnet.net> for
669 helping me find this bug.
671 4. Fixed the treatment for an interrupt signal when dump access
672 the remote tape through RSH. Thanks to Christian Weisgerber
673 <naddy@mips.rhein-neckar.de> for providing the patch.
675 5. Fixed a bug which was causing dump/restore to display
676 garbage characters instead of the remote host name.
678 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
679 =====================================================================
681 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
682 <kukuk@suse.de> for submitting the patch.
684 2. Fix for the exit code when using the size estimate option of
685 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
686 submitting the patch.
688 3. Handle EINTR in atomical reads/writes in dump, which was causing
689 dump to fail on some systems. Thanks to Eric Jergensen
690 <eric@dvns.com> for reporting the bug and submitting the patch.
692 4. Handle more than 16 characters for the device names in dumpdates.
693 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
694 tracking down the problem and proposing the solution.
696 5. Fixed a bug in dump which prevented the creation of the
697 dumpdates file when doing a 0-level dump without already
698 having a dumpdates file. Thanks to Patrik Schindler
699 <poc@pocnet.net> for reporting the bug.
701 6. Changed the way dump 'S' flag reports the size estimate
702 from number of blocks into bytes (making it compatible
703 with the Solaris version, and simplifying things for
704 amanda users). Thanks to Jason L Tibbitts III
705 <tibbs@math.uh.edu> for reporting the bug.
707 7. Fixed a compatibility problem in linux/alpha dump tape format.
708 Now the linux/alpha dump are (again) compatible with the
709 other dump formats. But this breaks compatibility with
710 older dumps made on alpha. Thanks to Mike Tibor
711 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
713 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
714 ======================================================================
716 1. Added a '--enable-kerberos' to configure.
718 2. Added a 'S' option to dump which determines the amount of space
719 that is needed to perform the dump without actually doing it, similar
720 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
721 <cermak@ahab.rutgers.edu>.
723 3. Added a 'M' multi-volume option to dump and restore which enables
724 dumping to multiple files (useful when dumping to an ext2
725 partition to make several dump files in order to bypass the 2GB
726 file size limitation). The 'f' argument is treated as a prefix and
727 the output files will be named <prefix>001, <prefix>002 etc. With
728 the 'M' flag, restore automatically selects the right file without
729 asking to enter a new tape each time.
731 4. Fixed a memory leak which was causing dump to grow very big
732 (270MB when dumping a 10GB filesystem...). Thanks to Jason
733 Fearon <jasonf@netrider.org.au> for reporting the bug.
735 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
736 ======================================================================
738 1. Make configure test if the system glob routines support
739 extended syntax (ALTDIRFUNC). If not, use the internal glob
740 routines instead of system ones. Thanks to Bernhard Sadlowski
741 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
742 and helping me resolve this and other minor libc5 compiling
745 2. Fix a problem when dumping a ext2fs with the 'filetype'
746 feature enabled. Thanks to Patrick J. LoPresti
747 <patl@cag.lcs.mit.edu> for reporting the bug and to
748 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
750 3. Made the nodump flag work on directories. A directory which
751 has the nodump flag gets never dumped, regardless of its
754 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
755 which allows dump on an active ext3 filesystem. However, this
756 is a "quick and dirty" patch which enables backup of an ext3
757 filesystem through the ext2 compatibility (by ignoring the
758 NEEDS_RECOVERY bit). The journal file is not recognized and
759 it is dumped (it should not).
761 5. Test the superblock compatibility flags when dumping, in order
762 to be sure that we know how to deal with specific features.
764 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
765 ====================================================================
767 1. Use lchown instead of chown, fixing a possible security problem
768 when restoring symlinks (a malicious user could use this
769 to deliberately corrupt the ownership of important system files).
770 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
771 this and providing the patch.
773 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
774 ====================================================================
776 1. Put dump sources under CVS, added Id tags in all files so
777 one can use 'ident' on binary files.
779 2. Added the dump/restore version in the usage text so one can
780 easily verify the version he is using.
782 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
783 a va_start/va_end problem on linux-ppc (always call va_start
784 va_end in pairs each time we use a vararg function).
786 4. Added again the DT_* constants because old libc does not
787 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
788 for submitting the bug report.
790 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
791 this should enable dumping big (huge) filesystems.
793 6. Added the RSH environment variable in order to be able to
794 use a rsh replacement like ssh when doing remote backups (and
795 bypass the security limitations of rcmd). Now you can do remote
796 backups without being root (or making dump setuid root).
798 7. Modified again the way dumpdates works. For incremental dumps,
799 we need to read dumpdates even if we are not using 'u' option.
800 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
803 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
804 ===================================================================
806 1. Removed the 'k' flag from the restore 'about' text if kerberos
809 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
810 to them (fsetflags takes a char*, setflags an open fd!).
812 3. (f)setflags is called only if the flags aren't empty. If the
813 file is a special file, a warning is printed, because changing
814 flags implies opening the device. Normally, a special file
815 should not have any flag... (Debian bug #29775, patch provided
816 by Abhijit Dasgupta <abhijit@ans.net>).
818 4. Made possible to dump a file system not mentioned in /etc/fstab.
819 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
821 5. Changed the default behaviour to not create dumpdates
822 unless 'u' option is specified. Removed the old "debian-patch"
823 which provided the same thing. (Debian bug #38136, #33818).
825 6. Removed all those dump*announce, since they were getting old...
827 7. Added warning messages if dumpdates does not exist and
828 when an inferior level dump does not exist (except for a level 0
831 8. Debugged the glob calls in interactive mode: restore used a
832 dirent struct which was different from the /usr/include/dirent.h
833 one (this used to work, is it a glibc2 change?), so none of the
834 compat glob (which used /usr/include/dirent.h) or the system glob
835 worked. Restore use now the system dirent (and the system
836 DT_* constants), which are compatible with BSD ones.
838 9. Added a configure flag (--with-dumpdatespath) to specify
839 the location of dumpdates file. By default, it is
842 10. Added the "AUTHOR" and "AVAILABILITY" sections and
843 included the current date/version in man pages.
845 11. Corrected the estimation of remaining time when
846 the operator doesn't change the tapes quickly enough. This
847 was an old bug, I thought I corrected it, and discovered
848 that in fact it was corrected in two different places, so
849 the results canceled each other...
851 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
852 ===================================================================
854 1. Integrated multiple patches from RedHat, Debian and SuSE:
856 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
858 - several fixes in the man pages.
859 - update the default tape device to /dev/st0.
860 - many updates for Linux Alpha (byte ordering, size_t etc).
862 - use environment variable for TMPDIR (instead of /tmp).
863 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
864 - workaround egcs bug (RedHat bugs #4281 and #2989).
865 - wire $(OPT) throughout Makefile's.
867 2. Upgrade the dump revision to 1, making possible to dump filesystems
868 made with e2fsprogs-1.15 or newer. Nothing seems to break...
870 3. Fix some compile warnings, prototype all functions.
872 4. Use glibc err/glob instead of internal compatibility
873 routines (only if available).
875 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
876 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
878 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
879 ======================================================================
881 1. Integrated the changes from FreeBSD-3.1-RELEASE
882 (mostly bug fixes, buffer overruns, dump has now an "automatic
883 tape length calculation" flag, dump/restore can use kerberos now
884 (this is NOT tested), use environment variables for TAPE and
887 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
889 3. Corrected a bug in restore when using 'C' option with multi-volumes
890 tapes (files splited accros two tapes give "size changed" errors
893 4. Corrected the long standing bug when dumping multiple tapes.
894 This works for me, needs further testing.
896 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
897 ====================================================================
899 1. Dump now runs correctly on kernels 2.1.x
900 Fix made by Gerald Peters <gapeters@worldnet.att.net>
902 Changes between versions 0.4b2 and 0.4b3
903 ========================================
905 1. Use realpath() if available
909 Changes between versions 0.4b1 and 0.4b2
910 ========================================
912 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
915 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
917 3. Fixed configure.in to correctly handle --enable-rmt
919 Changes between versions 0.3 and 0.4b1
920 ======================================
922 1. Integrated the changes from 4.4BSD-Lite2
924 2. Integrated the patches from Debian and Red Hat
926 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
928 4. Changed dump to use the Ext2fs library to get block addresses. This
929 should solve the endianness problem on SparcLinux.
931 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
932 distribution's one) to use autoconf
934 6. Fixed a few minor bugs
936 Changes between versions 0.2e and 0.2f
937 ======================================
939 1. Added the creation of named pipes (fifos) in restore.
941 2. Added the -N flag in the restore manual page.
943 3. Added the file kernel.patch which contains the llseek() optimization
944 patch for 1.2.x kernels.
946 4. Fixed a bug in the restoration of symbolic links: owner and group were
949 5. Integrated some changes from FreeBSD 2.2.
951 6. Added a call to ftruncate() after restoring each file to restore
952 correctly files ending by a hole.
954 Changes between versions 0.2d and 0.2e
955 ======================================
957 1. Fixed a bug in the "set owner/mode" process. Every file was restored
958 with owner = root (0) and group = root/wheel/whatever (0).
960 Changes between versions 0.2c and 0.2d
961 ======================================
963 1. Dump is now able to backup 2GB+ filesystems.
965 2. Dump and restore can now be linked as static binaries.
967 Changes between versions 0.2b and 0.2c
968 ======================================
970 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
972 Changes between versions 0.2a and 0.2b
973 ======================================
975 1. Really fixed the bug that I should have corrected in 0.2a.
977 2. Enabled optimization again.
979 Changes between versions 0.2 and 0.2a
980 =====================================
982 1. Disabled the optimization during compilation.
984 Changes between versions 0.1 and 0.2
985 ====================================
987 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
988 the fs_type field (actually, I modified the file fstab.c to make it
989 use the mntent functions).
991 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
992 made by dump should be readable by the BSD restore and backups made
993 by the BSD dump should be readable by restore. Unfortunately, this
994 means that the dump format has changed between version 0.1 and version
997 3. Dump is now able to backup a subtree, it is no longer limited to whole
998 filesystems like the BSD version.
1000 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1003 Changes between versions 0.0 and 0.1
1004 ====================================
1006 1. Now create links rdump and rrestore during the `make install' step.
1008 2. Linux port specific bugs added to the manual pages
1010 3. Incorrect estimation of the number of tapes blocks fixed when doing
1011 incremental backups.
1013 4. Better ls-like format in restore in interactive mode.