1 $Id: CHANGES,v 1.182 2002/06/25 19:00:37 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 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
16 ==================================================================
18 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
19 non Linux clients were misinterpreted. The description of the problem
20 (incompatible numbering in Linux mtio opcodes) is documented at
21 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
22 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
23 providing an excellent, cross-platform replacement for rmt in his
26 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
27 which caused restore to fail when the inode maps were not entirely
28 contained on the first volume (when using really small volumes or
29 when dumping a huge number of inodes).
31 3. Fixed a problem in dump, where files in subdirectories of directories
32 that have the nodump flag set (or being excluded with -e/-E) were
33 sometimes incorrectly dumped (instead of being excluded). The original
34 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
35 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
36 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
37 help testing the patch.
39 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
40 which makes one able to pipe the output of dump, by the net, to
41 a remote CD-burner server.
43 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
44 and use GNU's symbolic syntax over rmt instead of numerical values
45 to assure multiple platform compatibility.
47 6. Documented the -d option in restore.
49 7. Added a -v (verbose) mode to dump. For now it just prints the number
50 of the inode being dumped, but this could evolve in future versions
51 to include interesting debugging output.
53 8. Added a -o flag to restore, which automatically restores the current
54 directory permissions in -i or -x mode, without asking the operator
55 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
56 and Peter Samuel <peters@e-smith.com>.
58 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
59 ====================================================================
61 1. Fixed a bug in the QFA feature which could have caused
62 a bad tape positionning on restore (causing extra delays in
63 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
65 2. Added a small note in the dump man page specifying that
66 there should be no white space between the option letter and
67 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
69 3. Made restore work with older versions of the readline library,
70 by checking in configure for several extended symbols. Restore
71 can now be compiled with a readline as old as the 2.0 release
72 (though it may be a good idea to upgrade it to a more recent
73 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
74 reporting the build failures.
76 4. Fixed a performance problem with the QFA file creation in
77 dump, which made unnecessary seeks on the tape slowing down
78 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
81 5. Removed the inclusion of some kernel headers in the dump
82 source, which prevented the compile in some kernel/glibc
83 headers/architecture combination. Thanks to Bdale Garbee
84 <bdale@gag.com> for reporting the bug.
86 6. Added the appropriate error message when dump fails to
87 open the output file for writing. Thanks to Amith Varghese
88 <amithv@yahoo.com> for reporting this bug.
90 7. Made restore able to understand large Solaris ufsdump tapes
91 (containing inodes bigger than 4194304). Sun have introduced
92 an "extension" to the dump tape format when dealing with
93 those inodes, which was uncorrectly handled by Linux restore.
94 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
95 providing a test case.
97 8. Added the -m parameter to dump which optimises the output for
98 inodes having been changed but not modified since the last dump
99 ('changed' and 'modified' have the meaning defined in stat(2)).
100 For those inodes, dump will save only the metadata, instead of
101 saving the entire inode contents. Inodes which are either
102 directories or have been modified since the last dump are saved
103 in a regular way. Uses of this flag must be consistent, meaning
104 that either every dump in an incremental dump set have the flag,
107 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
108 =======================================================================
110 1. Fixed behaviour of dump when exceeding resource limits
113 2. Added the -L flag to restore to allow the user to specify a
114 maximal allowed number of miscompares when using restore
115 with the -C option to check the backup.
117 3. Detailed the manual entry for the -N option of restore.
119 4. Added the -a flag to restore to make able doing unattended
120 restores in -i or -x mode (automatically walks through the
121 multiple dump volumes).
123 5. Extended the QFA mode to work with local files and/or
124 remote tapes and files. This way, restore can know in advance
125 the tape number and the offset for the inodes to extract and
126 can minimize the extraction time by seeking directly to the
129 6. Added the -A <archive> option to both dump and restore,
130 which makes dump to archive a dump table-of-contents in
131 the specified file to be used by restore to determine
132 whether a file is in the dump file that is being restored.
133 (the archive file syntax is also compatible with the
134 Solaris ufsdump generated one).
136 7. Small fix in restore making it able to read some (broken ?)
137 Solaris ufsdump tapes.
139 8. Fixed dump to correctly recognise the root filesystem when using
140 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
141 for reporting this bug.
143 9. Added the -P <file> option to restore to create a
144 Quick File Access file from an already made dump. Patch
145 contributed by Uwe Gohlke <uwe@ugsoft.de>.
147 10. Made restore compile and run on Solaris, making it a
148 possible replacement for the standard ufsrestore. Port was
149 contributed by Uwe Gohlke <uwe@ugsoft.de>.
151 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
152 =====================================================================
154 1. Added a set of backup scripts from Eugenio Diaz
155 <getnito@yahoo.com> in the examples section. It features
156 automatic (cron based) full and incremental dumping of
157 several filesystems on a separate filesystem.
159 2. Fixed a off-by-one miscalculation which disabled dumping
160 a one letter subdirectory. Thanks to Chris Siebenmann
161 <cks@utcc.utoronto.ca> for reporting the bug.
163 3. Fixed several restore bugs occuring when trying to
164 restore some missing files on the tape. Thanks to Chris
165 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
167 4. Fixed --with-ldopts configure argument passing, installing from
168 a separate object directory, makefile cleanups contributed
169 by <splite@purdue.edu>.
171 5. Fix a bug which could caused, in some conditions, the highest
172 number inode of a filesystem, to not be dumped. Many thanks
173 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
176 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
177 =======================================================================
179 1. Added a mini howto from Patrick Walsh in the examples
182 2. Minor man pages syntax corrections. Thanks to
183 Chris Danis <screechco@home.com> for reporting the bugs.
185 3. Added a script from David B. Peterson <dave@toppledwagon.com>
186 to the examples section. It features dumping several
187 filesystems to a remote tape drive upon ssh.
189 4. Added a patch provided by Richard Jones <rich@annexia.org>
190 which allows BRADEMAX (number of read errors tolerated by
191 dump) to be adjusted using the -I option.
193 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
194 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
197 6. Add the -l option to restore to specify if, when doing a
198 remote restore, the file used to access the backup is a
199 regular file (the defaults being a tape drive). Restore needs
200 to know this information when reading a remote compressed
201 dump. Previously, this information was autodetected, but
202 the autodetection code fails (with ioctl: Inappropriate ioctl
203 for device) when using a non Linux remote box. Thanks to
204 many users and especially to Eros Albertazzi
205 <eros@lamel.bo.cnr.it> for reporting this.
207 7. Found a workaroung for the dump deadlock problem (3 childs
208 stuck in pause(), father in read()). The workaround seems
209 to work for me and several beta-testers. If it doesn't work
210 for you, please report back.
212 8. Updated the RPM spec file (BuildPrereq, URL etc).
214 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
215 ========================================================================
217 1. Fixed the permissions of a newly created QFA file by dump.
219 2. Cleaned up the source of dump (the external variables
220 definition was a complete mess, making possible to have
223 3. Fixed restore to use the full tape volume path when doing
224 a compare (since it changes the working directory to the
225 filesystem being compared in the process).
227 4. Added the -q option to dump which makes dump abort
228 whenever it needs operator attention. It should be
229 easier to use dump in scripts with this option.
231 5. Detect the use of incompatible options to dump and
232 refuse them (like -a and -B options together).
234 6. Added bzip2 compression to dump/restore (use option -j level
235 to select it). Note that you will need a restore version
236 >= 0.4b24 in order to restore a bzip2 compressed dump.
237 The same warning as for the zlib compression applies:
238 the tape format of a bzip2 dump is not compatible with the
239 original BSD tape format.
241 7. Fixed a overflow problem in dump corrupting the dump when
242 very large files were encountered. Thanks to Vladimir Nadvornik
245 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
246 kernel buffer/page cache before starting the dump, helping
247 a bit those who use dump on mounted filesystems. Thanks to
248 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
251 9. Updated the RPM spec file following the RedHat changes
252 (dynamically linked binaries now in /usr/sbin etc).
254 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
255 which enables restore to recognise multi volume compressed dumps
256 done on CD. Included his scripts for dump (which pipe the dump
257 output directly into cdrecord) and restore. There is now
258 possible to to backups to CD on the fly!
260 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
261 ===================================================================
263 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
264 Marcus Meissner (Caldera International Security Dept.).
266 2. Implement the Sun rmt extended protocol. Patch submitted
267 by Ian Gordon <iangordon@users.sourceforge.net>.
269 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
270 the linux kernel header. This ensures that dump always has the
271 latest version of this file. Patch submitted by
272 Andreas Dilger <adilger@turbolinux.com>.
274 4. Report any filesystem present in either /etc/fstab with a
275 positive passno or /etc/dumpdates in dump -w output.
276 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
278 5. Fixed the looping problem in dump introduced in the
281 6. Changed the -B option of dump to limit the size of
282 _compressed_ output per volume if compression is on.
283 Patch contributed by Helmut Jarausch
284 <jarausch@igpm.rwth-aachen.de>. Note however that, since
285 it is impossible to predict the size of the compressed
286 data before doing the compression, the -B limit is a bit
289 7. Fixed a bug in reading the operator typed file/tape path for
290 the next volume in restore.
292 8. Implemented a "-F script" option to restore which permits the
293 user to specify a script which will be launched at the
294 beginning of each tape, useful for automatic programming of
295 tape changers for example. See the restore man page for the
296 script parameters and return codes.
298 9. Small fix for the QFA routines provided by Uwe Gohlke
299 <uwe@ugsoft.de>, and some recommendations for QFA uses in
302 10. Fixed the multivolume restoring where making a mistake
303 to the 'Mount next tape' prompt caused several blocks to
306 11. Enhanced the -e option of dump to take as a parameter a
307 comma separated list of inode numbers.
309 12. Added the -E option to dump which specify a file containing
310 inode numbers to exclude from the dump.
312 13. Fixed the compressed multi-volume dump + restore.
314 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
315 ==================================================================
317 1. Made dump report the number of blocks dumped per volume.
318 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
320 2. Fix a bug in dump where waiting too much at the 'change volumes'
321 question caused the volume to be rewritten. Thanks to
322 Graham Reed <greed@users.sourceforge.net> for reporting the
323 bug and providing a patch.
325 3. Added a compression option to dump, contributed by Jerry
326 Peters <gapeters@worldnet.att.net>.
328 WARNING: the compression works only on files/pipes and
329 drives supporting variable block size.
331 WARNING2: tapes created using the compression feature are
332 incompatible with the standard dump tape format, and a
333 version of restore >= 0.4b22 will be needed for extraction.
335 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
336 architectures. Thanks to Paul Slootman <paul@debian.org> for
337 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
340 5. Many cleanups (CPP defines, const char warnings, check of
341 ext2fs COMPAT flags, time_t cleanups, added libext2 version
342 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
344 6. Made --prefix option work in configure. All the install path
345 are now based on the configure parameters.
347 7. Added the Quick File Access mode in dump/restore, contributed
348 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
349 a file tape position for each inode, and this file is used by
350 restore (if called with parameter Q and the filename)
351 to directly position the tape at the file restore is currently
352 working on. This saves hours when restoring single files from
353 large backups, saves the tapes and the drive's head. Use
354 --enable-qfa option of configure to compile in the QFA support.
356 8. Added the possibility to dump several files and directories
357 in a single invocation of dump. Thanks to Uwe Gohlke
358 <uwe@ugsoft.de> for implementing this option.
360 9. Fixed the dumping and restoring of files with holes
361 (files ending with holes didn't get dumped or restored
364 10. Fixed a socket descriptor leak in dump, which leaved opened
365 3 file descriptors per dump process (and there is one dump
368 11. Fixed dump large file system support, by explicit use of
369 open64/lseek64/etc functions (explicit use needed because
370 e2fsprogs libraries don't behave well when compiled with
371 FILE_OFFSET_BITS=64).
373 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
374 ======================================================================
376 1. Fixed some bugs in the dump -w|-W logic introduced by
377 the previous version. Thanks to Andreas Dilger
378 <adilger@turbolinux.com> for his help on this one.
380 2. Fixed again a compilation problem when using old e2fs
381 headers (filesystem label related). Thanks to many users
382 who reported this stupid error.
384 3. Fixed a build problem on old lib5 based systems dealing with
385 _PATH_MNTTAB being undefined. Thanks to John Adams
386 <johna@onevista.com> for reporting the bug.
388 4. Improved the error detection in restore when creating
389 the temporary files in TMPDIR. Now restore will corectly
390 report a 'no space left on device' error instead of
391 strange errors which could imply an archive corruption.
392 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
393 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
395 5. Added the throughput information in dump's progression
396 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
399 6. Use libext2fs's inode scanning routines, which are particularly
400 robust when dealing with errors (especially when having some
401 bad blocks on the filesystem). Many thanks to Colin
402 <colin@horizon.com> for reporting the bug, investigating
403 the issues, finding a workaround, writing the patch and
404 fully testing it... (of course, if this patch will break
405 anything, he is to blame for :-)).
407 7. Made dump and restore LFS aware. Dump can dump a filesystem
408 containing large files, generate a large file on output and
409 restore can restore them. This feature is not enabled by
410 default, use --enable-largefile option of configure to enable
411 it (you need to have a LFS aware glibc though). Thanks to
412 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
413 and to Theodore T'so <tytso@valinux.com> for his always
416 8. Made dump ask upon a tape write error if it should rewrite
417 the current volume (assume this is a bad tape) or if it should
418 assume an end-of-tape condition (useful for tape drives which
419 behaves badly at the end of the tape). Thanks to Andreas
420 Dilger <adilger@turbolinux.com> for the suggestion.
422 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
423 =======================================================================
425 1. Fixed a small compilation problem due to a change
426 in the definintion of the struct sigaction in
427 glibc 2.0 / libc5. Thanks to Gunther Schlegel
428 <schlegel@riege.de> for reporting the bug and to
429 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
432 2. Modified the label and uuid analysis in order to be
433 self-contained (independant of kernel/libc headers). This
434 should fix the compile with older kernel/libc headers and
435 will preserve the functionality. Thanks to Bernhard Erdmann
436 <bernhard.erdmann@gmx.de> for reporting the bug.
438 3. The 'exclude inode' option, if applied to a directory
439 inode, excludes now correctly all the leaves of this
440 directory. Thanks to John R. Dennison
441 <gerdesas@users.sourceforge.net> for reporting the bug.
443 4. Fixed the '-e' option to disable the possibility
444 to exclude the root inode (which causes the dumps to
445 be unreadable by restore). Prevented array overflow
446 when multiple -e options are used.
448 5. Fixed dump to correctly interpret a filesystem argument
449 which is a mountpoint and it is not an absolute path
450 (as specified in the fstab). Thanks to Bernhard R. Erdmann
451 <be@berdmann.de> for reporting the bug.
453 6. Made dump able to backup files larger than 2 GB. Note that
454 dump still doesn't cope with files larger than 4 GB.
456 7. Restore the real uid and gid before invoking an external
457 RSH program (big hole when dump or restore is suid root!).
459 8. Get the values for the default tape device and the location
460 of fstab file from the system headers. Thanks to
461 Andreas Dilger <adilger@turbolinux.com> for the patch.
463 9. Made dump -w|-W report all recognized filesystems
464 present in either /etc/fstab or /etc/dumpdates, and present
465 the list in the same order as in fstab file. Thanks
466 to Andreas Dilger <adilger@turbolinux.com> for the patch.
468 10. Made dump's -a (automatic end of tape detection) the
469 default. Specifying one of -B, -c, -d or -s options will
470 override the EOM detection. Thanks to Andreas Dilger
471 <adilger@turbolinux.com> for the patch.
473 11. Save the ext2 filesystem label into the dump volume label.
474 Specifying a volume label on the command line (-L option)
475 will override this feature. Thanks to Andreas Dilger
476 <adilger@turbolinux.com> for the patch.
478 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
479 =====================================================================
481 1. Fixed the signal handling in dump (which I broke in 0.4b17)
482 which was causing several strange dump failures (dump
483 hanged or segmentation faults etc).
485 2. Specified the default blocksize in the dump man page.
487 3. Changed two info messages of restore to be written on stdout
488 instead of stderr in order to leave stderr only for errors
489 or warnings. Thanks to Stephen Atwell
490 <satwell@urbana.css.mot.com> for the suggestion.
492 4. Corrected an off by one calculation which prevented
493 dumping a directory having a 1 character name length.
494 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
495 for reporting the bug.
497 5. Reinforce dump in order to allow only level 0 and no
498 -u option when dumping a subdirectory, and document
499 this in the man page. Thanks to Bernhard Erdmann
500 <bernhard.erdmann@gmx.de> for reporting the bug.
502 6. Fixed a small harmless bug in restore which caused
503 in some conditions hard links to be restored several
504 times (therefore generation some warning messages).
505 Thanks to Kenneth Porter <shiva@well.com> for
508 7. Updated the RPM spec file to the latest RedHat version,
509 providing FHS packaging and other cosmetic changes.
510 You will need RPM version >= 3.0.5 in order to build the RPM.
512 8. Updated the configure script to check for libtermcap
513 before checking for libreadline, because we need this
514 library in order to compile the readline support.
516 9. Made dump understand the LABEL= and UUID= notation
517 both in /etc/fstab and on the command line. Note that
518 you will need the /proc filesystem in order to use
519 these notations. Thanks to Erik Troan <ewt@redhat.com>
520 for providing the patch.
522 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
523 ===================================================================
525 1. Fixed a potential buffer overflow in restore. Thanks
526 to Stan Bubrouski <satan@fastdial.net> for reporting
529 2. Fixed a readline-related bug which prevented
530 'cat DUMPFILE | restore -i -f -' from working. Thanks
531 to Charles Karney <karney@users.sourceforge.net>
534 3. Changed a few "panic" into "exit", causing restore to
535 be more stable against some attacks (like the last one
536 reported on Bugtraq, although the last version of restore
537 was not vulnerable - just dumped core). Thanks to
538 Andreas Hasenack <andreas@conectiva.com.br> for reporting
541 4. Removed the suid-root bit on dump and restore in the
542 default build (and generated RPMs). It should be safer
543 now. Users who need the suid-root capabilities in order
544 to make network backups should read first the man page
545 of dump and enable the suid bit manually.
547 5. Added -ltermcap to the compile parameters for restore
548 when using readline support, in order to make the compile
549 process work on some systems (SuSE ?). Thanks to
550 Patrik Schindler <poc@pocnet.net> for reporting the bug.
552 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
553 ====================================================================
555 1. The -F script is called now *only* at the end of a tape,
556 not at the end of the dump. Thanks to Christian Haul
557 <haul@informatik.tu-darmstadt.de> for the bug report.
559 Normally, the device name and the current volume number
560 are passed on the command line of the script. However,
561 if you want the old style script (with no arguments
562 passed on the command line), you can enable it in
563 configure with the --enable-oldstylefscript.
565 2. Use posix signal handling to preserve dump functionality
566 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
569 3. Made the exit code of restore in case of a 'C'ompare
570 command reflect the result of the compare. An exit status
571 of 0 means the dump archive is correct, 1 means tape errors,
572 2 means that some files were modified. Thanks to Kenneth Porter
573 <shiva@well.com> for the suggestion.
575 4. Made (finally) quotes work in interactive restore.
577 5. Small fixes in order to allow dump to compile with a
578 really old e2fsprogs version. Thanks to Ian Zimmerman
579 <itz@speakeasy.org> for the bug report.
581 6. Add GNU readline capabilities to interactive restore.
582 Use configure's --enable-readline flag to enable this feature.
583 Thanks to Patrik Schindler <poc@pocnet.net> for the
586 7. Do the compare on the fly in restore 'C' mode (this will
587 allow not to exhaust the available /tmp space when
588 ccmparing large files). Thanks to Kenneth Porter
589 <shiva@well.com> for the suggestion.
591 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
592 ====================================================================
594 1. Fixed some several duplicate 'DUMP: DUMP:' in the
597 2. Corrected the estimation of blocks to dump. Note that
598 this is still wrong for large directory inodes, due
599 to the size of a BSD directory entry on the tape
600 (estimation is based on the size of the ext2 directory,
601 which is smaller because it doesn't contain the null
602 character at the end).
604 3. Made dump report the total number of MB written to
605 tape at the end of run. Thanks to W. Reilly Cooley
606 <wcooley@nakedape.cc> for the patch.
608 4. Added the -X option to restore giving the possibility
609 to read the names of the files to be extracted/listed
610 from a text file (in addition of the command line).
611 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
614 5. Added the device name and the current volume number
615 as arguments to the end of tape script (dump -F option).
617 6. Made the multi-volume dump work again (after having
618 broken it in 0.4b15).
620 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
621 ===================================================================
623 1. Added a prompt command in interactive restore mode. Thanks
624 to Andreas Dilger <adilger@home.com> for the patch.
626 2. Fixed a buffer overflow problem in dump (caused by
627 not checking the size of the filesystem parameter).
628 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
629 reporting this on Bugtraq (and to several dump users
630 who forwarded me his mail).
632 3. Added the '-F script' option to dump in order to
633 launch a script at the end of each tape (to be used
634 with a tape changer, or to notify the sysadmin by
637 4. Fixed a bug in restore compare code caused by the changes
640 5. Fixed the treatment of options using the old BSD syntax
641 in both dump and restore.
643 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
644 =======================================================================
646 1. Fixed a bug in dump which may have caused invalid deleted
647 directories to be dumped out if they were deleted after the
648 mapping pass. This could occure on active filesystem and lead
649 to heap corruption (causing dump malfunction in many possible ways).
650 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
651 bug and submitting the patch.
653 2. Fixed the handling of the filesystem parameter in dump. A
654 '/mnt/disk' parameter caused the disk contents to be dumped,
655 but a '/mnt/disk/' parameter caused the mountpoint directory
656 to be dumped (generally an empty directory).
658 3. Improved the output of dump in order to tell which directory
659 it is currently dumping (when dumping a subtree).
661 4. Added the '-e' exclude inode option to dump. Thanks to
662 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
664 5. Added a REPORTING-BUGS file in order to provide a guide
665 on how to correctly report a bug in dump/restore.
667 6. Fixed a restore problem when restoring a hard link to an inode
668 having the immutable or append-only attribute set. Thanks to
669 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
671 7. Fixed a compatibility problem between dumps made on little
672 endian machines (the format was correct) and big endian
673 machines (the format was incorrect). This fix break the
674 compatibility with the older dumps made on big endian
675 machines (sparc, m86k, ppc etc). For the first time in
676 linux dump's history (I believe), the dumps made by *BSD,
677 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
678 so interchangeable. Thanks to Rob Cermak
679 <cermak@ahab.rutgers.edu> for submitting the bug and
680 helping me test the fix.
682 8. Fixed the way dump reports the remaining percent/time, if
683 the number of blocks actually dumped exceeds the estimated
684 number of blocks. Thanks to Jean-Paul van der Jagt
685 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
687 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
688 ======================================================================
690 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
691 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
693 2. Removed the requirement to build the RPM as root from the
694 spec file. Thanks to Christian Weisgerber
695 <naddy@mips.rhein-neckar.de> for submitting this.
697 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
698 causing dump to block when dumping really huge filesystems.
699 Many thanks to Patrik Schindler <poc@pocnet.net> for
700 helping me find this bug.
702 4. Fixed the treatment for an interrupt signal when dump access
703 the remote tape through RSH. Thanks to Christian Weisgerber
704 <naddy@mips.rhein-neckar.de> for providing the patch.
706 5. Fixed a bug which was causing dump/restore to display
707 garbage characters instead of the remote host name.
709 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
710 =====================================================================
712 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
713 <kukuk@suse.de> for submitting the patch.
715 2. Fix for the exit code when using the size estimate option of
716 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
717 submitting the patch.
719 3. Handle EINTR in atomical reads/writes in dump, which was causing
720 dump to fail on some systems. Thanks to Eric Jergensen
721 <eric@dvns.com> for reporting the bug and submitting the patch.
723 4. Handle more than 16 characters for the device names in dumpdates.
724 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
725 tracking down the problem and proposing the solution.
727 5. Fixed a bug in dump which prevented the creation of the
728 dumpdates file when doing a 0-level dump without already
729 having a dumpdates file. Thanks to Patrik Schindler
730 <poc@pocnet.net> for reporting the bug.
732 6. Changed the way dump 'S' flag reports the size estimate
733 from number of blocks into bytes (making it compatible
734 with the Solaris version, and simplifying things for
735 amanda users). Thanks to Jason L Tibbitts III
736 <tibbs@math.uh.edu> for reporting the bug.
738 7. Fixed a compatibility problem in linux/alpha dump tape format.
739 Now the linux/alpha dump are (again) compatible with the
740 other dump formats. But this breaks compatibility with
741 older dumps made on alpha. Thanks to Mike Tibor
742 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
744 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
745 ======================================================================
747 1. Added a '--enable-kerberos' to configure.
749 2. Added a 'S' option to dump which determines the amount of space
750 that is needed to perform the dump without actually doing it, similar
751 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
752 <cermak@ahab.rutgers.edu>.
754 3. Added a 'M' multi-volume option to dump and restore which enables
755 dumping to multiple files (useful when dumping to an ext2
756 partition to make several dump files in order to bypass the 2GB
757 file size limitation). The 'f' argument is treated as a prefix and
758 the output files will be named <prefix>001, <prefix>002 etc. With
759 the 'M' flag, restore automatically selects the right file without
760 asking to enter a new tape each time.
762 4. Fixed a memory leak which was causing dump to grow very big
763 (270MB when dumping a 10GB filesystem...). Thanks to Jason
764 Fearon <jasonf@netrider.org.au> for reporting the bug.
766 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
767 ======================================================================
769 1. Make configure test if the system glob routines support
770 extended syntax (ALTDIRFUNC). If not, use the internal glob
771 routines instead of system ones. Thanks to Bernhard Sadlowski
772 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
773 and helping me resolve this and other minor libc5 compiling
776 2. Fix a problem when dumping a ext2fs with the 'filetype'
777 feature enabled. Thanks to Patrick J. LoPresti
778 <patl@cag.lcs.mit.edu> for reporting the bug and to
779 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
781 3. Made the nodump flag work on directories. A directory which
782 has the nodump flag gets never dumped, regardless of its
785 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
786 which allows dump on an active ext3 filesystem. However, this
787 is a "quick and dirty" patch which enables backup of an ext3
788 filesystem through the ext2 compatibility (by ignoring the
789 NEEDS_RECOVERY bit). The journal file is not recognized and
790 it is dumped (it should not).
792 5. Test the superblock compatibility flags when dumping, in order
793 to be sure that we know how to deal with specific features.
795 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
796 ====================================================================
798 1. Use lchown instead of chown, fixing a possible security problem
799 when restoring symlinks (a malicious user could use this
800 to deliberately corrupt the ownership of important system files).
801 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
802 this and providing the patch.
804 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
805 ====================================================================
807 1. Put dump sources under CVS, added Id tags in all files so
808 one can use 'ident' on binary files.
810 2. Added the dump/restore version in the usage text so one can
811 easily verify the version he is using.
813 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
814 a va_start/va_end problem on linux-ppc (always call va_start
815 va_end in pairs each time we use a vararg function).
817 4. Added again the DT_* constants because old libc does not
818 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
819 for submitting the bug report.
821 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
822 this should enable dumping big (huge) filesystems.
824 6. Added the RSH environment variable in order to be able to
825 use a rsh replacement like ssh when doing remote backups (and
826 bypass the security limitations of rcmd). Now you can do remote
827 backups without being root (or making dump setuid root).
829 7. Modified again the way dumpdates works. For incremental dumps,
830 we need to read dumpdates even if we are not using 'u' option.
831 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
834 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
835 ===================================================================
837 1. Removed the 'k' flag from the restore 'about' text if kerberos
840 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
841 to them (fsetflags takes a char*, setflags an open fd!).
843 3. (f)setflags is called only if the flags aren't empty. If the
844 file is a special file, a warning is printed, because changing
845 flags implies opening the device. Normally, a special file
846 should not have any flag... (Debian bug #29775, patch provided
847 by Abhijit Dasgupta <abhijit@ans.net>).
849 4. Made possible to dump a file system not mentioned in /etc/fstab.
850 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
852 5. Changed the default behaviour to not create dumpdates
853 unless 'u' option is specified. Removed the old "debian-patch"
854 which provided the same thing. (Debian bug #38136, #33818).
856 6. Removed all those dump*announce, since they were getting old...
858 7. Added warning messages if dumpdates does not exist and
859 when an inferior level dump does not exist (except for a level 0
862 8. Debugged the glob calls in interactive mode: restore used a
863 dirent struct which was different from the /usr/include/dirent.h
864 one (this used to work, is it a glibc2 change?), so none of the
865 compat glob (which used /usr/include/dirent.h) or the system glob
866 worked. Restore use now the system dirent (and the system
867 DT_* constants), which are compatible with BSD ones.
869 9. Added a configure flag (--with-dumpdatespath) to specify
870 the location of dumpdates file. By default, it is
873 10. Added the "AUTHOR" and "AVAILABILITY" sections and
874 included the current date/version in man pages.
876 11. Corrected the estimation of remaining time when
877 the operator doesn't change the tapes quickly enough. This
878 was an old bug, I thought I corrected it, and discovered
879 that in fact it was corrected in two different places, so
880 the results canceled each other...
882 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
883 ===================================================================
885 1. Integrated multiple patches from RedHat, Debian and SuSE:
887 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
889 - several fixes in the man pages.
890 - update the default tape device to /dev/st0.
891 - many updates for Linux Alpha (byte ordering, size_t etc).
893 - use environment variable for TMPDIR (instead of /tmp).
894 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
895 - workaround egcs bug (RedHat bugs #4281 and #2989).
896 - wire $(OPT) throughout Makefile's.
898 2. Upgrade the dump revision to 1, making possible to dump filesystems
899 made with e2fsprogs-1.15 or newer. Nothing seems to break...
901 3. Fix some compile warnings, prototype all functions.
903 4. Use glibc err/glob instead of internal compatibility
904 routines (only if available).
906 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
907 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
909 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
910 ======================================================================
912 1. Integrated the changes from FreeBSD-3.1-RELEASE
913 (mostly bug fixes, buffer overruns, dump has now an "automatic
914 tape length calculation" flag, dump/restore can use kerberos now
915 (this is NOT tested), use environment variables for TAPE and
918 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
920 3. Corrected a bug in restore when using 'C' option with multi-volumes
921 tapes (files splited accros two tapes give "size changed" errors
924 4. Corrected the long standing bug when dumping multiple tapes.
925 This works for me, needs further testing.
927 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
928 ====================================================================
930 1. Dump now runs correctly on kernels 2.1.x
931 Fix made by Gerald Peters <gapeters@worldnet.att.net>
933 Changes between versions 0.4b2 and 0.4b3
934 ========================================
936 1. Use realpath() if available
940 Changes between versions 0.4b1 and 0.4b2
941 ========================================
943 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
946 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
948 3. Fixed configure.in to correctly handle --enable-rmt
950 Changes between versions 0.3 and 0.4b1
951 ======================================
953 1. Integrated the changes from 4.4BSD-Lite2
955 2. Integrated the patches from Debian and Red Hat
957 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
959 4. Changed dump to use the Ext2fs library to get block addresses. This
960 should solve the endianness problem on SparcLinux.
962 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
963 distribution's one) to use autoconf
965 6. Fixed a few minor bugs
967 Changes between versions 0.2e and 0.2f
968 ======================================
970 1. Added the creation of named pipes (fifos) in restore.
972 2. Added the -N flag in the restore manual page.
974 3. Added the file kernel.patch which contains the llseek() optimization
975 patch for 1.2.x kernels.
977 4. Fixed a bug in the restoration of symbolic links: owner and group were
980 5. Integrated some changes from FreeBSD 2.2.
982 6. Added a call to ftruncate() after restoring each file to restore
983 correctly files ending by a hole.
985 Changes between versions 0.2d and 0.2e
986 ======================================
988 1. Fixed a bug in the "set owner/mode" process. Every file was restored
989 with owner = root (0) and group = root/wheel/whatever (0).
991 Changes between versions 0.2c and 0.2d
992 ======================================
994 1. Dump is now able to backup 2GB+ filesystems.
996 2. Dump and restore can now be linked as static binaries.
998 Changes between versions 0.2b and 0.2c
999 ======================================
1001 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1003 Changes between versions 0.2a and 0.2b
1004 ======================================
1006 1. Really fixed the bug that I should have corrected in 0.2a.
1008 2. Enabled optimization again.
1010 Changes between versions 0.2 and 0.2a
1011 =====================================
1013 1. Disabled the optimization during compilation.
1015 Changes between versions 0.1 and 0.2
1016 ====================================
1018 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1019 the fs_type field (actually, I modified the file fstab.c to make it
1020 use the mntent functions).
1022 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1023 made by dump should be readable by the BSD restore and backups made
1024 by the BSD dump should be readable by restore. Unfortunately, this
1025 means that the dump format has changed between version 0.1 and version
1028 3. Dump is now able to backup a subtree, it is no longer limited to whole
1029 filesystems like the BSD version.
1031 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1034 Changes between versions 0.0 and 0.1
1035 ====================================
1037 1. Now create links rdump and rrestore during the `make install' step.
1039 2. Linux port specific bugs added to the manual pages
1041 3. Incorrect estimation of the number of tapes blocks fixed when doing
1042 incremental backups.
1044 4. Better ls-like format in restore in interactive mode.