1 $Id: CHANGES,v 1.180 2002/06/08 07:19:28 stelian Exp $
3 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
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 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
28 which makes one able to pipe the output of dump, by the net, to
29 a remote CD-burner server.
31 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
32 and use GNU's symbolic syntax over rmt instead of numerical values
33 to assure multiple platform compatibility.
35 6. Documented the -d option in restore.
37 7. Added a -v (verbose) mode to dump. For now it just prints the number
38 of the inode being dumped, but this could evolve in future versions
39 to include interesting debugging output.
41 8. Added a -o flag to restore, which automatically restores the current
42 directory permissions in -i or -x mode, without asking the operator
43 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
44 and Peter Samuel <peters@e-smith.com>.
46 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
47 ====================================================================
49 1. Fixed a bug in the QFA feature which could have caused
50 a bad tape positionning on restore (causing extra delays in
51 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
53 2. Added a small note in the dump man page specifying that
54 there should be no white space between the option letter and
55 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
57 3. Made restore work with older versions of the readline library,
58 by checking in configure for several extended symbols. Restore
59 can now be compiled with a readline as old as the 2.0 release
60 (though it may be a good idea to upgrade it to a more recent
61 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
62 reporting the build failures.
64 4. Fixed a performance problem with the QFA file creation in
65 dump, which made unnecessary seeks on the tape slowing down
66 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
69 5. Removed the inclusion of some kernel headers in the dump
70 source, which prevented the compile in some kernel/glibc
71 headers/architecture combination. Thanks to Bdale Garbee
72 <bdale@gag.com> for reporting the bug.
74 6. Added the appropriate error message when dump fails to
75 open the output file for writing. Thanks to Amith Varghese
76 <amithv@yahoo.com> for reporting this bug.
78 7. Made restore able to understand large Solaris ufsdump tapes
79 (containing inodes bigger than 4194304). Sun have introduced
80 an "extension" to the dump tape format when dealing with
81 those inodes, which was uncorrectly handled by Linux restore.
82 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
83 providing a test case.
85 8. Added the -m parameter to dump which optimises the output for
86 inodes having been changed but not modified since the last dump
87 ('changed' and 'modified' have the meaning defined in stat(2)).
88 For those inodes, dump will save only the metadata, instead of
89 saving the entire inode contents. Inodes which are either
90 directories or have been modified since the last dump are saved
91 in a regular way. Uses of this flag must be consistent, meaning
92 that either every dump in an incremental dump set have the flag,
95 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
96 =======================================================================
98 1. Fixed behaviour of dump when exceeding resource limits
101 2. Added the -L flag to restore to allow the user to specify a
102 maximal allowed number of miscompares when using restore
103 with the -C option to check the backup.
105 3. Detailed the manual entry for the -N option of restore.
107 4. Added the -a flag to restore to make able doing unattended
108 restores in -i or -x mode (automatically walks through the
109 multiple dump volumes).
111 5. Extended the QFA mode to work with local files and/or
112 remote tapes and files. This way, restore can know in advance
113 the tape number and the offset for the inodes to extract and
114 can minimize the extraction time by seeking directly to the
117 6. Added the -A <archive> option to both dump and restore,
118 which makes dump to archive a dump table-of-contents in
119 the specified file to be used by restore to determine
120 whether a file is in the dump file that is being restored.
121 (the archive file syntax is also compatible with the
122 Solaris ufsdump generated one).
124 7. Small fix in restore making it able to read some (broken ?)
125 Solaris ufsdump tapes.
127 8. Fixed dump to correctly recognise the root filesystem when using
128 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
129 for reporting this bug.
131 9. Added the -P <file> option to restore to create a
132 Quick File Access file from an already made dump. Patch
133 contributed by Uwe Gohlke <uwe@ugsoft.de>.
135 10. Made restore compile and run on Solaris, making it a
136 possible replacement for the standard ufsrestore. Port was
137 contributed by Uwe Gohlke <uwe@ugsoft.de>.
139 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
140 =====================================================================
142 1. Added a set of backup scripts from Eugenio Diaz
143 <getnito@yahoo.com> in the examples section. It features
144 automatic (cron based) full and incremental dumping of
145 several filesystems on a separate filesystem.
147 2. Fixed a off-by-one miscalculation which disabled dumping
148 a one letter subdirectory. Thanks to Chris Siebenmann
149 <cks@utcc.utoronto.ca> for reporting the bug.
151 3. Fixed several restore bugs occuring when trying to
152 restore some missing files on the tape. Thanks to Chris
153 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
155 4. Fixed --with-ldopts configure argument passing, installing from
156 a separate object directory, makefile cleanups contributed
157 by <splite@purdue.edu>.
159 5. Fix a bug which could caused, in some conditions, the highest
160 number inode of a filesystem, to not be dumped. Many thanks
161 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
164 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
165 =======================================================================
167 1. Added a mini howto from Patrick Walsh in the examples
170 2. Minor man pages syntax corrections. Thanks to
171 Chris Danis <screechco@home.com> for reporting the bugs.
173 3. Added a script from David B. Peterson <dave@toppledwagon.com>
174 to the examples section. It features dumping several
175 filesystems to a remote tape drive upon ssh.
177 4. Added a patch provided by Richard Jones <rich@annexia.org>
178 which allows BRADEMAX (number of read errors tolerated by
179 dump) to be adjusted using the -I option.
181 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
182 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
185 6. Add the -l option to restore to specify if, when doing a
186 remote restore, the file used to access the backup is a
187 regular file (the defaults being a tape drive). Restore needs
188 to know this information when reading a remote compressed
189 dump. Previously, this information was autodetected, but
190 the autodetection code fails (with ioctl: Inappropriate ioctl
191 for device) when using a non Linux remote box. Thanks to
192 many users and especially to Eros Albertazzi
193 <eros@lamel.bo.cnr.it> for reporting this.
195 7. Found a workaroung for the dump deadlock problem (3 childs
196 stuck in pause(), father in read()). The workaround seems
197 to work for me and several beta-testers. If it doesn't work
198 for you, please report back.
200 8. Updated the RPM spec file (BuildPrereq, URL etc).
202 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
203 ========================================================================
205 1. Fixed the permissions of a newly created QFA file by dump.
207 2. Cleaned up the source of dump (the external variables
208 definition was a complete mess, making possible to have
211 3. Fixed restore to use the full tape volume path when doing
212 a compare (since it changes the working directory to the
213 filesystem being compared in the process).
215 4. Added the -q option to dump which makes dump abort
216 whenever it needs operator attention. It should be
217 easier to use dump in scripts with this option.
219 5. Detect the use of incompatible options to dump and
220 refuse them (like -a and -B options together).
222 6. Added bzip2 compression to dump/restore (use option -j level
223 to select it). Note that you will need a restore version
224 >= 0.4b24 in order to restore a bzip2 compressed dump.
225 The same warning as for the zlib compression applies:
226 the tape format of a bzip2 dump is not compatible with the
227 original BSD tape format.
229 7. Fixed a overflow problem in dump corrupting the dump when
230 very large files were encountered. Thanks to Vladimir Nadvornik
233 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
234 kernel buffer/page cache before starting the dump, helping
235 a bit those who use dump on mounted filesystems. Thanks to
236 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
239 9. Updated the RPM spec file following the RedHat changes
240 (dynamically linked binaries now in /usr/sbin etc).
242 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
243 which enables restore to recognise multi volume compressed dumps
244 done on CD. Included his scripts for dump (which pipe the dump
245 output directly into cdrecord) and restore. There is now
246 possible to to backups to CD on the fly!
248 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
249 ===================================================================
251 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
252 Marcus Meissner (Caldera International Security Dept.).
254 2. Implement the Sun rmt extended protocol. Patch submitted
255 by Ian Gordon <iangordon@users.sourceforge.net>.
257 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
258 the linux kernel header. This ensures that dump always has the
259 latest version of this file. Patch submitted by
260 Andreas Dilger <adilger@turbolinux.com>.
262 4. Report any filesystem present in either /etc/fstab with a
263 positive passno or /etc/dumpdates in dump -w output.
264 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
266 5. Fixed the looping problem in dump introduced in the
269 6. Changed the -B option of dump to limit the size of
270 _compressed_ output per volume if compression is on.
271 Patch contributed by Helmut Jarausch
272 <jarausch@igpm.rwth-aachen.de>. Note however that, since
273 it is impossible to predict the size of the compressed
274 data before doing the compression, the -B limit is a bit
277 7. Fixed a bug in reading the operator typed file/tape path for
278 the next volume in restore.
280 8. Implemented a "-F script" option to restore which permits the
281 user to specify a script which will be launched at the
282 beginning of each tape, useful for automatic programming of
283 tape changers for example. See the restore man page for the
284 script parameters and return codes.
286 9. Small fix for the QFA routines provided by Uwe Gohlke
287 <uwe@ugsoft.de>, and some recommendations for QFA uses in
290 10. Fixed the multivolume restoring where making a mistake
291 to the 'Mount next tape' prompt caused several blocks to
294 11. Enhanced the -e option of dump to take as a parameter a
295 comma separated list of inode numbers.
297 12. Added the -E option to dump which specify a file containing
298 inode numbers to exclude from the dump.
300 13. Fixed the compressed multi-volume dump + restore.
302 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
303 ==================================================================
305 1. Made dump report the number of blocks dumped per volume.
306 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
308 2. Fix a bug in dump where waiting too much at the 'change volumes'
309 question caused the volume to be rewritten. Thanks to
310 Graham Reed <greed@users.sourceforge.net> for reporting the
311 bug and providing a patch.
313 3. Added a compression option to dump, contributed by Jerry
314 Peters <gapeters@worldnet.att.net>.
316 WARNING: the compression works only on files/pipes and
317 drives supporting variable block size.
319 WARNING2: tapes created using the compression feature are
320 incompatible with the standard dump tape format, and a
321 version of restore >= 0.4b22 will be needed for extraction.
323 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
324 architectures. Thanks to Paul Slootman <paul@debian.org> for
325 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
328 5. Many cleanups (CPP defines, const char warnings, check of
329 ext2fs COMPAT flags, time_t cleanups, added libext2 version
330 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
332 6. Made --prefix option work in configure. All the install path
333 are now based on the configure parameters.
335 7. Added the Quick File Access mode in dump/restore, contributed
336 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
337 a file tape position for each inode, and this file is used by
338 restore (if called with parameter Q and the filename)
339 to directly position the tape at the file restore is currently
340 working on. This saves hours when restoring single files from
341 large backups, saves the tapes and the drive's head. Use
342 --enable-qfa option of configure to compile in the QFA support.
344 8. Added the possibility to dump several files and directories
345 in a single invocation of dump. Thanks to Uwe Gohlke
346 <uwe@ugsoft.de> for implementing this option.
348 9. Fixed the dumping and restoring of files with holes
349 (files ending with holes didn't get dumped or restored
352 10. Fixed a socket descriptor leak in dump, which leaved opened
353 3 file descriptors per dump process (and there is one dump
356 11. Fixed dump large file system support, by explicit use of
357 open64/lseek64/etc functions (explicit use needed because
358 e2fsprogs libraries don't behave well when compiled with
359 FILE_OFFSET_BITS=64).
361 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
362 ======================================================================
364 1. Fixed some bugs in the dump -w|-W logic introduced by
365 the previous version. Thanks to Andreas Dilger
366 <adilger@turbolinux.com> for his help on this one.
368 2. Fixed again a compilation problem when using old e2fs
369 headers (filesystem label related). Thanks to many users
370 who reported this stupid error.
372 3. Fixed a build problem on old lib5 based systems dealing with
373 _PATH_MNTTAB being undefined. Thanks to John Adams
374 <johna@onevista.com> for reporting the bug.
376 4. Improved the error detection in restore when creating
377 the temporary files in TMPDIR. Now restore will corectly
378 report a 'no space left on device' error instead of
379 strange errors which could imply an archive corruption.
380 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
381 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
383 5. Added the throughput information in dump's progression
384 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
387 6. Use libext2fs's inode scanning routines, which are particularly
388 robust when dealing with errors (especially when having some
389 bad blocks on the filesystem). Many thanks to Colin
390 <colin@horizon.com> for reporting the bug, investigating
391 the issues, finding a workaround, writing the patch and
392 fully testing it... (of course, if this patch will break
393 anything, he is to blame for :-)).
395 7. Made dump and restore LFS aware. Dump can dump a filesystem
396 containing large files, generate a large file on output and
397 restore can restore them. This feature is not enabled by
398 default, use --enable-largefile option of configure to enable
399 it (you need to have a LFS aware glibc though). Thanks to
400 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
401 and to Theodore T'so <tytso@valinux.com> for his always
404 8. Made dump ask upon a tape write error if it should rewrite
405 the current volume (assume this is a bad tape) or if it should
406 assume an end-of-tape condition (useful for tape drives which
407 behaves badly at the end of the tape). Thanks to Andreas
408 Dilger <adilger@turbolinux.com> for the suggestion.
410 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
411 =======================================================================
413 1. Fixed a small compilation problem due to a change
414 in the definintion of the struct sigaction in
415 glibc 2.0 / libc5. Thanks to Gunther Schlegel
416 <schlegel@riege.de> for reporting the bug and to
417 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
420 2. Modified the label and uuid analysis in order to be
421 self-contained (independant of kernel/libc headers). This
422 should fix the compile with older kernel/libc headers and
423 will preserve the functionality. Thanks to Bernhard Erdmann
424 <bernhard.erdmann@gmx.de> for reporting the bug.
426 3. The 'exclude inode' option, if applied to a directory
427 inode, excludes now correctly all the leaves of this
428 directory. Thanks to John R. Dennison
429 <gerdesas@users.sourceforge.net> for reporting the bug.
431 4. Fixed the '-e' option to disable the possibility
432 to exclude the root inode (which causes the dumps to
433 be unreadable by restore). Prevented array overflow
434 when multiple -e options are used.
436 5. Fixed dump to correctly interpret a filesystem argument
437 which is a mountpoint and it is not an absolute path
438 (as specified in the fstab). Thanks to Bernhard R. Erdmann
439 <be@berdmann.de> for reporting the bug.
441 6. Made dump able to backup files larger than 2 GB. Note that
442 dump still doesn't cope with files larger than 4 GB.
444 7. Restore the real uid and gid before invoking an external
445 RSH program (big hole when dump or restore is suid root!).
447 8. Get the values for the default tape device and the location
448 of fstab file from the system headers. Thanks to
449 Andreas Dilger <adilger@turbolinux.com> for the patch.
451 9. Made dump -w|-W report all recognized filesystems
452 present in either /etc/fstab or /etc/dumpdates, and present
453 the list in the same order as in fstab file. Thanks
454 to Andreas Dilger <adilger@turbolinux.com> for the patch.
456 10. Made dump's -a (automatic end of tape detection) the
457 default. Specifying one of -B, -c, -d or -s options will
458 override the EOM detection. Thanks to Andreas Dilger
459 <adilger@turbolinux.com> for the patch.
461 11. Save the ext2 filesystem label into the dump volume label.
462 Specifying a volume label on the command line (-L option)
463 will override this feature. Thanks to Andreas Dilger
464 <adilger@turbolinux.com> for the patch.
466 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
467 =====================================================================
469 1. Fixed the signal handling in dump (which I broke in 0.4b17)
470 which was causing several strange dump failures (dump
471 hanged or segmentation faults etc).
473 2. Specified the default blocksize in the dump man page.
475 3. Changed two info messages of restore to be written on stdout
476 instead of stderr in order to leave stderr only for errors
477 or warnings. Thanks to Stephen Atwell
478 <satwell@urbana.css.mot.com> for the suggestion.
480 4. Corrected an off by one calculation which prevented
481 dumping a directory having a 1 character name length.
482 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
483 for reporting the bug.
485 5. Reinforce dump in order to allow only level 0 and no
486 -u option when dumping a subdirectory, and document
487 this in the man page. Thanks to Bernhard Erdmann
488 <bernhard.erdmann@gmx.de> for reporting the bug.
490 6. Fixed a small harmless bug in restore which caused
491 in some conditions hard links to be restored several
492 times (therefore generation some warning messages).
493 Thanks to Kenneth Porter <shiva@well.com> for
496 7. Updated the RPM spec file to the latest RedHat version,
497 providing FHS packaging and other cosmetic changes.
498 You will need RPM version >= 3.0.5 in order to build the RPM.
500 8. Updated the configure script to check for libtermcap
501 before checking for libreadline, because we need this
502 library in order to compile the readline support.
504 9. Made dump understand the LABEL= and UUID= notation
505 both in /etc/fstab and on the command line. Note that
506 you will need the /proc filesystem in order to use
507 these notations. Thanks to Erik Troan <ewt@redhat.com>
508 for providing the patch.
510 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
511 ===================================================================
513 1. Fixed a potential buffer overflow in restore. Thanks
514 to Stan Bubrouski <satan@fastdial.net> for reporting
517 2. Fixed a readline-related bug which prevented
518 'cat DUMPFILE | restore -i -f -' from working. Thanks
519 to Charles Karney <karney@users.sourceforge.net>
522 3. Changed a few "panic" into "exit", causing restore to
523 be more stable against some attacks (like the last one
524 reported on Bugtraq, although the last version of restore
525 was not vulnerable - just dumped core). Thanks to
526 Andreas Hasenack <andreas@conectiva.com.br> for reporting
529 4. Removed the suid-root bit on dump and restore in the
530 default build (and generated RPMs). It should be safer
531 now. Users who need the suid-root capabilities in order
532 to make network backups should read first the man page
533 of dump and enable the suid bit manually.
535 5. Added -ltermcap to the compile parameters for restore
536 when using readline support, in order to make the compile
537 process work on some systems (SuSE ?). Thanks to
538 Patrik Schindler <poc@pocnet.net> for reporting the bug.
540 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
541 ====================================================================
543 1. The -F script is called now *only* at the end of a tape,
544 not at the end of the dump. Thanks to Christian Haul
545 <haul@informatik.tu-darmstadt.de> for the bug report.
547 Normally, the device name and the current volume number
548 are passed on the command line of the script. However,
549 if you want the old style script (with no arguments
550 passed on the command line), you can enable it in
551 configure with the --enable-oldstylefscript.
553 2. Use posix signal handling to preserve dump functionality
554 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
557 3. Made the exit code of restore in case of a 'C'ompare
558 command reflect the result of the compare. An exit status
559 of 0 means the dump archive is correct, 1 means tape errors,
560 2 means that some files were modified. Thanks to Kenneth Porter
561 <shiva@well.com> for the suggestion.
563 4. Made (finally) quotes work in interactive restore.
565 5. Small fixes in order to allow dump to compile with a
566 really old e2fsprogs version. Thanks to Ian Zimmerman
567 <itz@speakeasy.org> for the bug report.
569 6. Add GNU readline capabilities to interactive restore.
570 Use configure's --enable-readline flag to enable this feature.
571 Thanks to Patrik Schindler <poc@pocnet.net> for the
574 7. Do the compare on the fly in restore 'C' mode (this will
575 allow not to exhaust the available /tmp space when
576 ccmparing large files). Thanks to Kenneth Porter
577 <shiva@well.com> for the suggestion.
579 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
580 ====================================================================
582 1. Fixed some several duplicate 'DUMP: DUMP:' in the
585 2. Corrected the estimation of blocks to dump. Note that
586 this is still wrong for large directory inodes, due
587 to the size of a BSD directory entry on the tape
588 (estimation is based on the size of the ext2 directory,
589 which is smaller because it doesn't contain the null
590 character at the end).
592 3. Made dump report the total number of MB written to
593 tape at the end of run. Thanks to W. Reilly Cooley
594 <wcooley@nakedape.cc> for the patch.
596 4. Added the -X option to restore giving the possibility
597 to read the names of the files to be extracted/listed
598 from a text file (in addition of the command line).
599 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
602 5. Added the device name and the current volume number
603 as arguments to the end of tape script (dump -F option).
605 6. Made the multi-volume dump work again (after having
606 broken it in 0.4b15).
608 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
609 ===================================================================
611 1. Added a prompt command in interactive restore mode. Thanks
612 to Andreas Dilger <adilger@home.com> for the patch.
614 2. Fixed a buffer overflow problem in dump (caused by
615 not checking the size of the filesystem parameter).
616 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
617 reporting this on Bugtraq (and to several dump users
618 who forwarded me his mail).
620 3. Added the '-F script' option to dump in order to
621 launch a script at the end of each tape (to be used
622 with a tape changer, or to notify the sysadmin by
625 4. Fixed a bug in restore compare code caused by the changes
628 5. Fixed the treatment of options using the old BSD syntax
629 in both dump and restore.
631 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
632 =======================================================================
634 1. Fixed a bug in dump which may have caused invalid deleted
635 directories to be dumped out if they were deleted after the
636 mapping pass. This could occure on active filesystem and lead
637 to heap corruption (causing dump malfunction in many possible ways).
638 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
639 bug and submitting the patch.
641 2. Fixed the handling of the filesystem parameter in dump. A
642 '/mnt/disk' parameter caused the disk contents to be dumped,
643 but a '/mnt/disk/' parameter caused the mountpoint directory
644 to be dumped (generally an empty directory).
646 3. Improved the output of dump in order to tell which directory
647 it is currently dumping (when dumping a subtree).
649 4. Added the '-e' exclude inode option to dump. Thanks to
650 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
652 5. Added a REPORTING-BUGS file in order to provide a guide
653 on how to correctly report a bug in dump/restore.
655 6. Fixed a restore problem when restoring a hard link to an inode
656 having the immutable or append-only attribute set. Thanks to
657 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
659 7. Fixed a compatibility problem between dumps made on little
660 endian machines (the format was correct) and big endian
661 machines (the format was incorrect). This fix break the
662 compatibility with the older dumps made on big endian
663 machines (sparc, m86k, ppc etc). For the first time in
664 linux dump's history (I believe), the dumps made by *BSD,
665 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
666 so interchangeable. Thanks to Rob Cermak
667 <cermak@ahab.rutgers.edu> for submitting the bug and
668 helping me test the fix.
670 8. Fixed the way dump reports the remaining percent/time, if
671 the number of blocks actually dumped exceeds the estimated
672 number of blocks. Thanks to Jean-Paul van der Jagt
673 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
675 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
676 ======================================================================
678 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
679 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
681 2. Removed the requirement to build the RPM as root from the
682 spec file. Thanks to Christian Weisgerber
683 <naddy@mips.rhein-neckar.de> for submitting this.
685 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
686 causing dump to block when dumping really huge filesystems.
687 Many thanks to Patrik Schindler <poc@pocnet.net> for
688 helping me find this bug.
690 4. Fixed the treatment for an interrupt signal when dump access
691 the remote tape through RSH. Thanks to Christian Weisgerber
692 <naddy@mips.rhein-neckar.de> for providing the patch.
694 5. Fixed a bug which was causing dump/restore to display
695 garbage characters instead of the remote host name.
697 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
698 =====================================================================
700 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
701 <kukuk@suse.de> for submitting the patch.
703 2. Fix for the exit code when using the size estimate option of
704 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
705 submitting the patch.
707 3. Handle EINTR in atomical reads/writes in dump, which was causing
708 dump to fail on some systems. Thanks to Eric Jergensen
709 <eric@dvns.com> for reporting the bug and submitting the patch.
711 4. Handle more than 16 characters for the device names in dumpdates.
712 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
713 tracking down the problem and proposing the solution.
715 5. Fixed a bug in dump which prevented the creation of the
716 dumpdates file when doing a 0-level dump without already
717 having a dumpdates file. Thanks to Patrik Schindler
718 <poc@pocnet.net> for reporting the bug.
720 6. Changed the way dump 'S' flag reports the size estimate
721 from number of blocks into bytes (making it compatible
722 with the Solaris version, and simplifying things for
723 amanda users). Thanks to Jason L Tibbitts III
724 <tibbs@math.uh.edu> for reporting the bug.
726 7. Fixed a compatibility problem in linux/alpha dump tape format.
727 Now the linux/alpha dump are (again) compatible with the
728 other dump formats. But this breaks compatibility with
729 older dumps made on alpha. Thanks to Mike Tibor
730 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
732 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
733 ======================================================================
735 1. Added a '--enable-kerberos' to configure.
737 2. Added a 'S' option to dump which determines the amount of space
738 that is needed to perform the dump without actually doing it, similar
739 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
740 <cermak@ahab.rutgers.edu>.
742 3. Added a 'M' multi-volume option to dump and restore which enables
743 dumping to multiple files (useful when dumping to an ext2
744 partition to make several dump files in order to bypass the 2GB
745 file size limitation). The 'f' argument is treated as a prefix and
746 the output files will be named <prefix>001, <prefix>002 etc. With
747 the 'M' flag, restore automatically selects the right file without
748 asking to enter a new tape each time.
750 4. Fixed a memory leak which was causing dump to grow very big
751 (270MB when dumping a 10GB filesystem...). Thanks to Jason
752 Fearon <jasonf@netrider.org.au> for reporting the bug.
754 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
755 ======================================================================
757 1. Make configure test if the system glob routines support
758 extended syntax (ALTDIRFUNC). If not, use the internal glob
759 routines instead of system ones. Thanks to Bernhard Sadlowski
760 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
761 and helping me resolve this and other minor libc5 compiling
764 2. Fix a problem when dumping a ext2fs with the 'filetype'
765 feature enabled. Thanks to Patrick J. LoPresti
766 <patl@cag.lcs.mit.edu> for reporting the bug and to
767 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
769 3. Made the nodump flag work on directories. A directory which
770 has the nodump flag gets never dumped, regardless of its
773 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
774 which allows dump on an active ext3 filesystem. However, this
775 is a "quick and dirty" patch which enables backup of an ext3
776 filesystem through the ext2 compatibility (by ignoring the
777 NEEDS_RECOVERY bit). The journal file is not recognized and
778 it is dumped (it should not).
780 5. Test the superblock compatibility flags when dumping, in order
781 to be sure that we know how to deal with specific features.
783 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
784 ====================================================================
786 1. Use lchown instead of chown, fixing a possible security problem
787 when restoring symlinks (a malicious user could use this
788 to deliberately corrupt the ownership of important system files).
789 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
790 this and providing the patch.
792 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
793 ====================================================================
795 1. Put dump sources under CVS, added Id tags in all files so
796 one can use 'ident' on binary files.
798 2. Added the dump/restore version in the usage text so one can
799 easily verify the version he is using.
801 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
802 a va_start/va_end problem on linux-ppc (always call va_start
803 va_end in pairs each time we use a vararg function).
805 4. Added again the DT_* constants because old libc does not
806 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
807 for submitting the bug report.
809 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
810 this should enable dumping big (huge) filesystems.
812 6. Added the RSH environment variable in order to be able to
813 use a rsh replacement like ssh when doing remote backups (and
814 bypass the security limitations of rcmd). Now you can do remote
815 backups without being root (or making dump setuid root).
817 7. Modified again the way dumpdates works. For incremental dumps,
818 we need to read dumpdates even if we are not using 'u' option.
819 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
822 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
823 ===================================================================
825 1. Removed the 'k' flag from the restore 'about' text if kerberos
828 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
829 to them (fsetflags takes a char*, setflags an open fd!).
831 3. (f)setflags is called only if the flags aren't empty. If the
832 file is a special file, a warning is printed, because changing
833 flags implies opening the device. Normally, a special file
834 should not have any flag... (Debian bug #29775, patch provided
835 by Abhijit Dasgupta <abhijit@ans.net>).
837 4. Made possible to dump a file system not mentioned in /etc/fstab.
838 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
840 5. Changed the default behaviour to not create dumpdates
841 unless 'u' option is specified. Removed the old "debian-patch"
842 which provided the same thing. (Debian bug #38136, #33818).
844 6. Removed all those dump*announce, since they were getting old...
846 7. Added warning messages if dumpdates does not exist and
847 when an inferior level dump does not exist (except for a level 0
850 8. Debugged the glob calls in interactive mode: restore used a
851 dirent struct which was different from the /usr/include/dirent.h
852 one (this used to work, is it a glibc2 change?), so none of the
853 compat glob (which used /usr/include/dirent.h) or the system glob
854 worked. Restore use now the system dirent (and the system
855 DT_* constants), which are compatible with BSD ones.
857 9. Added a configure flag (--with-dumpdatespath) to specify
858 the location of dumpdates file. By default, it is
861 10. Added the "AUTHOR" and "AVAILABILITY" sections and
862 included the current date/version in man pages.
864 11. Corrected the estimation of remaining time when
865 the operator doesn't change the tapes quickly enough. This
866 was an old bug, I thought I corrected it, and discovered
867 that in fact it was corrected in two different places, so
868 the results canceled each other...
870 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
871 ===================================================================
873 1. Integrated multiple patches from RedHat, Debian and SuSE:
875 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
877 - several fixes in the man pages.
878 - update the default tape device to /dev/st0.
879 - many updates for Linux Alpha (byte ordering, size_t etc).
881 - use environment variable for TMPDIR (instead of /tmp).
882 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
883 - workaround egcs bug (RedHat bugs #4281 and #2989).
884 - wire $(OPT) throughout Makefile's.
886 2. Upgrade the dump revision to 1, making possible to dump filesystems
887 made with e2fsprogs-1.15 or newer. Nothing seems to break...
889 3. Fix some compile warnings, prototype all functions.
891 4. Use glibc err/glob instead of internal compatibility
892 routines (only if available).
894 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
895 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
897 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
898 ======================================================================
900 1. Integrated the changes from FreeBSD-3.1-RELEASE
901 (mostly bug fixes, buffer overruns, dump has now an "automatic
902 tape length calculation" flag, dump/restore can use kerberos now
903 (this is NOT tested), use environment variables for TAPE and
906 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
908 3. Corrected a bug in restore when using 'C' option with multi-volumes
909 tapes (files splited accros two tapes give "size changed" errors
912 4. Corrected the long standing bug when dumping multiple tapes.
913 This works for me, needs further testing.
915 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
916 ====================================================================
918 1. Dump now runs correctly on kernels 2.1.x
919 Fix made by Gerald Peters <gapeters@worldnet.att.net>
921 Changes between versions 0.4b2 and 0.4b3
922 ========================================
924 1. Use realpath() if available
928 Changes between versions 0.4b1 and 0.4b2
929 ========================================
931 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
934 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
936 3. Fixed configure.in to correctly handle --enable-rmt
938 Changes between versions 0.3 and 0.4b1
939 ======================================
941 1. Integrated the changes from 4.4BSD-Lite2
943 2. Integrated the patches from Debian and Red Hat
945 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
947 4. Changed dump to use the Ext2fs library to get block addresses. This
948 should solve the endianness problem on SparcLinux.
950 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
951 distribution's one) to use autoconf
953 6. Fixed a few minor bugs
955 Changes between versions 0.2e and 0.2f
956 ======================================
958 1. Added the creation of named pipes (fifos) in restore.
960 2. Added the -N flag in the restore manual page.
962 3. Added the file kernel.patch which contains the llseek() optimization
963 patch for 1.2.x kernels.
965 4. Fixed a bug in the restoration of symbolic links: owner and group were
968 5. Integrated some changes from FreeBSD 2.2.
970 6. Added a call to ftruncate() after restoring each file to restore
971 correctly files ending by a hole.
973 Changes between versions 0.2d and 0.2e
974 ======================================
976 1. Fixed a bug in the "set owner/mode" process. Every file was restored
977 with owner = root (0) and group = root/wheel/whatever (0).
979 Changes between versions 0.2c and 0.2d
980 ======================================
982 1. Dump is now able to backup 2GB+ filesystems.
984 2. Dump and restore can now be linked as static binaries.
986 Changes between versions 0.2b and 0.2c
987 ======================================
989 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
991 Changes between versions 0.2a and 0.2b
992 ======================================
994 1. Really fixed the bug that I should have corrected in 0.2a.
996 2. Enabled optimization again.
998 Changes between versions 0.2 and 0.2a
999 =====================================
1001 1. Disabled the optimization during compilation.
1003 Changes between versions 0.1 and 0.2
1004 ====================================
1006 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1007 the fs_type field (actually, I modified the file fstab.c to make it
1008 use the mntent functions).
1010 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1011 made by dump should be readable by the BSD restore and backups made
1012 by the BSD dump should be readable by restore. Unfortunately, this
1013 means that the dump format has changed between version 0.1 and version
1016 3. Dump is now able to backup a subtree, it is no longer limited to whole
1017 filesystems like the BSD version.
1019 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1022 Changes between versions 0.0 and 0.1
1023 ====================================
1025 1. Now create links rdump and rrestore during the `make install' step.
1027 2. Linux port specific bugs added to the manual pages
1029 3. Incorrect estimation of the number of tapes blocks fixed when doing
1030 incremental backups.
1032 4. Better ls-like format in restore in interactive mode.