1 $Id: CHANGES,v 1.173 2002/04/15 11:57:27 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 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
15 ====================================================================
17 1. Fixed a bug in the QFA feature which could have caused
18 a bad tape positionning on restore (causing extra delays in
19 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
21 2. Added a small note in the dump man page specifying that
22 there should be no white space between the option letter and
23 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
25 3. Made restore work with older versions of the readline library,
26 by checking in configure for several extended symbols. Restore
27 can now be compiled with a readline as old as the 2.0 release
28 (though it may be a good idea to upgrade it to a more recent
29 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
30 reporting the build failures.
32 4. Fixed a performance problem with the QFA file creation in
33 dump, which made unnecessary seeks on the tape slowing down
34 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
37 5. Removed the inclusion of some kernel headers in the dump
38 source, which prevented the compile in some kernel/glibc
39 headers/architecture combination. Thanks to Bdale Garbee
40 <bdale@gag.com> for reporting the bug.
42 6. Added the appropriate error message when dump fails to
43 open the output file for writing. Thanks to Amith Varghese
44 <amithv@yahoo.com> for reporting this bug.
46 7. Made restore able to understand large Solaris ufsdump tapes
47 (containing inodes bigger than 4194304). Sun have introduced
48 an "extension" to the dump tape format when dealing with
49 those inodes, which was uncorrectly handled by Linux restore.
50 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
51 providing a test case.
53 8. Added the -m parameter to dump which optimises the output for
54 inodes having been changed but not modified since the last dump
55 ('changed' and 'modified' have the meaning defined in stat(2)).
56 For those inodes, dump will save only the metadata, instead of
57 saving the entire inode contents. Inodes which are either
58 directories or have been modified since the last dump are saved
59 in a regular way. Uses of this flag must be consistent, meaning
60 that either every dump in an incremental dump set have the flag,
63 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
64 =======================================================================
66 1. Fixed behaviour of dump when exceeding resource limits
69 2. Added the -L flag to restore to allow the user to specify a
70 maximal allowed number of miscompares when using restore
71 with the -C option to check the backup.
73 3. Detailed the manual entry for the -N option of restore.
75 4. Added the -a flag to restore to make able doing unattended
76 restores in -i or -x mode (automatically walks through the
77 multiple dump volumes).
79 5. Extended the QFA mode to work with local files and/or
80 remote tapes and files. This way, restore can know in advance
81 the tape number and the offset for the inodes to extract and
82 can minimize the extraction time by seeking directly to the
85 6. Added the -A <archive> option to both dump and restore,
86 which makes dump to archive a dump table-of-contents in
87 the specified file to be used by restore to determine
88 whether a file is in the dump file that is being restored.
89 (the archive file syntax is also compatible with the
90 Solaris ufsdump generated one).
92 7. Small fix in restore making it able to read some (broken ?)
93 Solaris ufsdump tapes.
95 8. Fixed dump to correctly recognise the root filesystem when using
96 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
97 for reporting this bug.
99 9. Added the -P <file> option to restore to create a
100 Quick File Access file from an already made dump. Patch
101 contributed by Uwe Gohlke <uwe@ugsoft.de>.
103 10. Made restore compile and run on Solaris, making it a
104 possible replacement for the standard ufsrestore. Port was
105 contributed by Uwe Gohlke <uwe@ugsoft.de>.
107 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
108 =====================================================================
110 1. Added a set of backup scripts from Eugenio Diaz
111 <getnito@yahoo.com> in the examples section. It features
112 automatic (cron based) full and incremental dumping of
113 several filesystems on a separate filesystem.
115 2. Fixed a off-by-one miscalculation which disabled dumping
116 a one letter subdirectory. Thanks to Chris Siebenmann
117 <cks@utcc.utoronto.ca> for reporting the bug.
119 3. Fixed several restore bugs occuring when trying to
120 restore some missing files on the tape. Thanks to Chris
121 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
123 4. Fixed --with-ldopts configure argument passing, installing from
124 a separate object directory, makefile cleanups contributed
125 by <splite@purdue.edu>.
127 5. Fix a bug which could caused, in some conditions, the highest
128 number inode of a filesystem, to not be dumped. Many thanks
129 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
132 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
133 =======================================================================
135 1. Added a mini howto from Patrick Walsh in the examples
138 2. Minor man pages syntax corrections. Thanks to
139 Chris Danis <screechco@home.com> for reporting the bugs.
141 3. Added a script from David B. Peterson <dave@toppledwagon.com>
142 to the examples section. It features dumping several
143 filesystems to a remote tape drive upon ssh.
145 4. Added a patch provided by Richard Jones <rich@annexia.org>
146 which allows BRADEMAX (number of read errors tolerated by
147 dump) to be adjusted using the -I option.
149 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
150 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
153 6. Add the -l option to restore to specify if, when doing a
154 remote restore, the file used to access the backup is a
155 regular file (the defaults being a tape drive). Restore needs
156 to know this information when reading a remote compressed
157 dump. Previously, this information was autodetected, but
158 the autodetection code fails (with ioctl: Inappropriate ioctl
159 for device) when using a non Linux remote box. Thanks to
160 many users and especially to Eros Albertazzi
161 <eros@lamel.bo.cnr.it> for reporting this.
163 7. Found a workaroung for the dump deadlock problem (3 childs
164 stuck in pause(), father in read()). The workaround seems
165 to work for me and several beta-testers. If it doesn't work
166 for you, please report back.
168 8. Updated the RPM spec file (BuildPrereq, URL etc).
170 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
171 ========================================================================
173 1. Fixed the permissions of a newly created QFA file by dump.
175 2. Cleaned up the source of dump (the external variables
176 definition was a complete mess, making possible to have
179 3. Fixed restore to use the full tape volume path when doing
180 a compare (since it changes the working directory to the
181 filesystem being compared in the process).
183 4. Added the -q option to dump which makes dump abort
184 whenever it needs operator attention. It should be
185 easier to use dump in scripts with this option.
187 5. Detect the use of incompatible options to dump and
188 refuse them (like -a and -B options together).
190 6. Added bzip2 compression to dump/restore (use option -j level
191 to select it). Note that you will need a restore version
192 >= 0.4b24 in order to restore a bzip2 compressed dump.
193 The same warning as for the zlib compression applies:
194 the tape format of a bzip2 dump is not compatible with the
195 original BSD tape format.
197 7. Fixed a overflow problem in dump corrupting the dump when
198 very large files were encountered. Thanks to Vladimir Nadvornik
201 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
202 kernel buffer/page cache before starting the dump, helping
203 a bit those who use dump on mounted filesystems. Thanks to
204 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
207 9. Updated the RPM spec file following the RedHat changes
208 (dynamically linked binaries now in /usr/sbin etc).
210 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
211 which enables restore to recognise multi volume compressed dumps
212 done on CD. Included his scripts for dump (which pipe the dump
213 output directly into cdrecord) and restore. There is now
214 possible to to backups to CD on the fly!
216 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
217 ===================================================================
219 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
220 Marcus Meissner (Caldera International Security Dept.).
222 2. Implement the Sun rmt extended protocol. Patch submitted
223 by Ian Gordon <iangordon@users.sourceforge.net>.
225 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
226 the linux kernel header. This ensures that dump always has the
227 latest version of this file. Patch submitted by
228 Andreas Dilger <adilger@turbolinux.com>.
230 4. Report any filesystem present in either /etc/fstab with a
231 positive passno or /etc/dumpdates in dump -w output.
232 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
234 5. Fixed the looping problem in dump introduced in the
237 6. Changed the -B option of dump to limit the size of
238 _compressed_ output per volume if compression is on.
239 Patch contributed by Helmut Jarausch
240 <jarausch@igpm.rwth-aachen.de>. Note however that, since
241 it is impossible to predict the size of the compressed
242 data before doing the compression, the -B limit is a bit
245 7. Fixed a bug in reading the operator typed file/tape path for
246 the next volume in restore.
248 8. Implemented a "-F script" option to restore which permits the
249 user to specify a script which will be launched at the
250 beginning of each tape, useful for automatic programming of
251 tape changers for example. See the restore man page for the
252 script parameters and return codes.
254 9. Small fix for the QFA routines provided by Uwe Gohlke
255 <uwe@ugsoft.de>, and some recommendations for QFA uses in
258 10. Fixed the multivolume restoring where making a mistake
259 to the 'Mount next tape' prompt caused several blocks to
262 11. Enhanced the -e option of dump to take as a parameter a
263 comma separated list of inode numbers.
265 12. Added the -E option to dump which specify a file containing
266 inode numbers to exclude from the dump.
268 13. Fixed the compressed multi-volume dump + restore.
270 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
271 ==================================================================
273 1. Made dump report the number of blocks dumped per volume.
274 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
276 2. Fix a bug in dump where waiting too much at the 'change volumes'
277 question caused the volume to be rewritten. Thanks to
278 Graham Reed <greed@users.sourceforge.net> for reporting the
279 bug and providing a patch.
281 3. Added a compression option to dump, contributed by Jerry
282 Peters <gapeters@worldnet.att.net>.
284 WARNING: the compression works only on files/pipes and
285 drives supporting variable block size.
287 WARNING2: tapes created using the compression feature are
288 incompatible with the standard dump tape format, and a
289 version of restore >= 0.4b22 will be needed for extraction.
291 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
292 architectures. Thanks to Paul Slootman <paul@debian.org> for
293 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
296 5. Many cleanups (CPP defines, const char warnings, check of
297 ext2fs COMPAT flags, time_t cleanups, added libext2 version
298 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
300 6. Made --prefix option work in configure. All the install path
301 are now based on the configure parameters.
303 7. Added the Quick File Access mode in dump/restore, contributed
304 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
305 a file tape position for each inode, and this file is used by
306 restore (if called with parameter Q and the filename)
307 to directly position the tape at the file restore is currently
308 working on. This saves hours when restoring single files from
309 large backups, saves the tapes and the drive's head. Use
310 --enable-qfa option of configure to compile in the QFA support.
312 8. Added the possibility to dump several files and directories
313 in a single invocation of dump. Thanks to Uwe Gohlke
314 <uwe@ugsoft.de> for implementing this option.
316 9. Fixed the dumping and restoring of files with holes
317 (files ending with holes didn't get dumped or restored
320 10. Fixed a socket descriptor leak in dump, which leaved opened
321 3 file descriptors per dump process (and there is one dump
324 11. Fixed dump large file system support, by explicit use of
325 open64/lseek64/etc functions (explicit use needed because
326 e2fsprogs libraries don't behave well when compiled with
327 FILE_OFFSET_BITS=64).
329 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
330 ======================================================================
332 1. Fixed some bugs in the dump -w|-W logic introduced by
333 the previous version. Thanks to Andreas Dilger
334 <adilger@turbolinux.com> for his help on this one.
336 2. Fixed again a compilation problem when using old e2fs
337 headers (filesystem label related). Thanks to many users
338 who reported this stupid error.
340 3. Fixed a build problem on old lib5 based systems dealing with
341 _PATH_MNTTAB being undefined. Thanks to John Adams
342 <johna@onevista.com> for reporting the bug.
344 4. Improved the error detection in restore when creating
345 the temporary files in TMPDIR. Now restore will corectly
346 report a 'no space left on device' error instead of
347 strange errors which could imply an archive corruption.
348 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
349 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
351 5. Added the throughput information in dump's progression
352 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
355 6. Use libext2fs's inode scanning routines, which are particularly
356 robust when dealing with errors (especially when having some
357 bad blocks on the filesystem). Many thanks to Colin
358 <colin@horizon.com> for reporting the bug, investigating
359 the issues, finding a workaround, writing the patch and
360 fully testing it... (of course, if this patch will break
361 anything, he is to blame for :-)).
363 7. Made dump and restore LFS aware. Dump can dump a filesystem
364 containing large files, generate a large file on output and
365 restore can restore them. This feature is not enabled by
366 default, use --enable-largefile option of configure to enable
367 it (you need to have a LFS aware glibc though). Thanks to
368 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
369 and to Theodore T'so <tytso@valinux.com> for his always
372 8. Made dump ask upon a tape write error if it should rewrite
373 the current volume (assume this is a bad tape) or if it should
374 assume an end-of-tape condition (useful for tape drives which
375 behaves badly at the end of the tape). Thanks to Andreas
376 Dilger <adilger@turbolinux.com> for the suggestion.
378 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
379 =======================================================================
381 1. Fixed a small compilation problem due to a change
382 in the definintion of the struct sigaction in
383 glibc 2.0 / libc5. Thanks to Gunther Schlegel
384 <schlegel@riege.de> for reporting the bug and to
385 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
388 2. Modified the label and uuid analysis in order to be
389 self-contained (independant of kernel/libc headers). This
390 should fix the compile with older kernel/libc headers and
391 will preserve the functionality. Thanks to Bernhard Erdmann
392 <bernhard.erdmann@gmx.de> for reporting the bug.
394 3. The 'exclude inode' option, if applied to a directory
395 inode, excludes now correctly all the leaves of this
396 directory. Thanks to John R. Dennison
397 <gerdesas@users.sourceforge.net> for reporting the bug.
399 4. Fixed the '-e' option to disable the possibility
400 to exclude the root inode (which causes the dumps to
401 be unreadable by restore). Prevented array overflow
402 when multiple -e options are used.
404 5. Fixed dump to correctly interpret a filesystem argument
405 which is a mountpoint and it is not an absolute path
406 (as specified in the fstab). Thanks to Bernhard R. Erdmann
407 <be@berdmann.de> for reporting the bug.
409 6. Made dump able to backup files larger than 2 GB. Note that
410 dump still doesn't cope with files larger than 4 GB.
412 7. Restore the real uid and gid before invoking an external
413 RSH program (big hole when dump or restore is suid root!).
415 8. Get the values for the default tape device and the location
416 of fstab file from the system headers. Thanks to
417 Andreas Dilger <adilger@turbolinux.com> for the patch.
419 9. Made dump -w|-W report all recognized filesystems
420 present in either /etc/fstab or /etc/dumpdates, and present
421 the list in the same order as in fstab file. Thanks
422 to Andreas Dilger <adilger@turbolinux.com> for the patch.
424 10. Made dump's -a (automatic end of tape detection) the
425 default. Specifying one of -B, -c, -d or -s options will
426 override the EOM detection. Thanks to Andreas Dilger
427 <adilger@turbolinux.com> for the patch.
429 11. Save the ext2 filesystem label into the dump volume label.
430 Specifying a volume label on the command line (-L option)
431 will override this feature. Thanks to Andreas Dilger
432 <adilger@turbolinux.com> for the patch.
434 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
435 =====================================================================
437 1. Fixed the signal handling in dump (which I broke in 0.4b17)
438 which was causing several strange dump failures (dump
439 hanged or segmentation faults etc).
441 2. Specified the default blocksize in the dump man page.
443 3. Changed two info messages of restore to be written on stdout
444 instead of stderr in order to leave stderr only for errors
445 or warnings. Thanks to Stephen Atwell
446 <satwell@urbana.css.mot.com> for the suggestion.
448 4. Corrected an off by one calculation which prevented
449 dumping a directory having a 1 character name length.
450 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
451 for reporting the bug.
453 5. Reinforce dump in order to allow only level 0 and no
454 -u option when dumping a subdirectory, and document
455 this in the man page. Thanks to Bernhard Erdmann
456 <bernhard.erdmann@gmx.de> for reporting the bug.
458 6. Fixed a small harmless bug in restore which caused
459 in some conditions hard links to be restored several
460 times (therefore generation some warning messages).
461 Thanks to Kenneth Porter <shiva@well.com> for
464 7. Updated the RPM spec file to the latest RedHat version,
465 providing FHS packaging and other cosmetic changes.
466 You will need RPM version >= 3.0.5 in order to build the RPM.
468 8. Updated the configure script to check for libtermcap
469 before checking for libreadline, because we need this
470 library in order to compile the readline support.
472 9. Made dump understand the LABEL= and UUID= notation
473 both in /etc/fstab and on the command line. Note that
474 you will need the /proc filesystem in order to use
475 these notations. Thanks to Erik Troan <ewt@redhat.com>
476 for providing the patch.
478 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
479 ===================================================================
481 1. Fixed a potential buffer overflow in restore. Thanks
482 to Stan Bubrouski <satan@fastdial.net> for reporting
485 2. Fixed a readline-related bug which prevented
486 'cat DUMPFILE | restore -i -f -' from working. Thanks
487 to Charles Karney <karney@users.sourceforge.net>
490 3. Changed a few "panic" into "exit", causing restore to
491 be more stable against some attacks (like the last one
492 reported on Bugtraq, although the last version of restore
493 was not vulnerable - just dumped core). Thanks to
494 Andreas Hasenack <andreas@conectiva.com.br> for reporting
497 4. Removed the suid-root bit on dump and restore in the
498 default build (and generated RPMs). It should be safer
499 now. Users who need the suid-root capabilities in order
500 to make network backups should read first the man page
501 of dump and enable the suid bit manually.
503 5. Added -ltermcap to the compile parameters for restore
504 when using readline support, in order to make the compile
505 process work on some systems (SuSE ?). Thanks to
506 Patrik Schindler <poc@pocnet.net> for reporting the bug.
508 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
509 ====================================================================
511 1. The -F script is called now *only* at the end of a tape,
512 not at the end of the dump. Thanks to Christian Haul
513 <haul@informatik.tu-darmstadt.de> for the bug report.
515 Normally, the device name and the current volume number
516 are passed on the command line of the script. However,
517 if you want the old style script (with no arguments
518 passed on the command line), you can enable it in
519 configure with the --enable-oldstylefscript.
521 2. Use posix signal handling to preserve dump functionality
522 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
525 3. Made the exit code of restore in case of a 'C'ompare
526 command reflect the result of the compare. An exit status
527 of 0 means the dump archive is correct, 1 means tape errors,
528 2 means that some files were modified. Thanks to Kenneth Porter
529 <shiva@well.com> for the suggestion.
531 4. Made (finally) quotes work in interactive restore.
533 5. Small fixes in order to allow dump to compile with a
534 really old e2fsprogs version. Thanks to Ian Zimmerman
535 <itz@speakeasy.org> for the bug report.
537 6. Add GNU readline capabilities to interactive restore.
538 Use configure's --enable-readline flag to enable this feature.
539 Thanks to Patrik Schindler <poc@pocnet.net> for the
542 7. Do the compare on the fly in restore 'C' mode (this will
543 allow not to exhaust the available /tmp space when
544 ccmparing large files). Thanks to Kenneth Porter
545 <shiva@well.com> for the suggestion.
547 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
548 ====================================================================
550 1. Fixed some several duplicate 'DUMP: DUMP:' in the
553 2. Corrected the estimation of blocks to dump. Note that
554 this is still wrong for large directory inodes, due
555 to the size of a BSD directory entry on the tape
556 (estimation is based on the size of the ext2 directory,
557 which is smaller because it doesn't contain the null
558 character at the end).
560 3. Made dump report the total number of MB written to
561 tape at the end of run. Thanks to W. Reilly Cooley
562 <wcooley@nakedape.cc> for the patch.
564 4. Added the -X option to restore giving the possibility
565 to read the names of the files to be extracted/listed
566 from a text file (in addition of the command line).
567 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
570 5. Added the device name and the current volume number
571 as arguments to the end of tape script (dump -F option).
573 6. Made the multi-volume dump work again (after having
574 broken it in 0.4b15).
576 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
577 ===================================================================
579 1. Added a prompt command in interactive restore mode. Thanks
580 to Andreas Dilger <adilger@home.com> for the patch.
582 2. Fixed a buffer overflow problem in dump (caused by
583 not checking the size of the filesystem parameter).
584 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
585 reporting this on Bugtraq (and to several dump users
586 who forwarded me his mail).
588 3. Added the '-F script' option to dump in order to
589 launch a script at the end of each tape (to be used
590 with a tape changer, or to notify the sysadmin by
593 4. Fixed a bug in restore compare code caused by the changes
596 5. Fixed the treatment of options using the old BSD syntax
597 in both dump and restore.
599 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
600 =======================================================================
602 1. Fixed a bug in dump which may have caused invalid deleted
603 directories to be dumped out if they were deleted after the
604 mapping pass. This could occure on active filesystem and lead
605 to heap corruption (causing dump malfunction in many possible ways).
606 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
607 bug and submitting the patch.
609 2. Fixed the handling of the filesystem parameter in dump. A
610 '/mnt/disk' parameter caused the disk contents to be dumped,
611 but a '/mnt/disk/' parameter caused the mountpoint directory
612 to be dumped (generally an empty directory).
614 3. Improved the output of dump in order to tell which directory
615 it is currently dumping (when dumping a subtree).
617 4. Added the '-e' exclude inode option to dump. Thanks to
618 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
620 5. Added a REPORTING-BUGS file in order to provide a guide
621 on how to correctly report a bug in dump/restore.
623 6. Fixed a restore problem when restoring a hard link to an inode
624 having the immutable or append-only attribute set. Thanks to
625 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
627 7. Fixed a compatibility problem between dumps made on little
628 endian machines (the format was correct) and big endian
629 machines (the format was incorrect). This fix break the
630 compatibility with the older dumps made on big endian
631 machines (sparc, m86k, ppc etc). For the first time in
632 linux dump's history (I believe), the dumps made by *BSD,
633 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
634 so interchangeable. Thanks to Rob Cermak
635 <cermak@ahab.rutgers.edu> for submitting the bug and
636 helping me test the fix.
638 8. Fixed the way dump reports the remaining percent/time, if
639 the number of blocks actually dumped exceeds the estimated
640 number of blocks. Thanks to Jean-Paul van der Jagt
641 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
643 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
644 ======================================================================
646 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
647 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
649 2. Removed the requirement to build the RPM as root from the
650 spec file. Thanks to Christian Weisgerber
651 <naddy@mips.rhein-neckar.de> for submitting this.
653 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
654 causing dump to block when dumping really huge filesystems.
655 Many thanks to Patrik Schindler <poc@pocnet.net> for
656 helping me find this bug.
658 4. Fixed the treatment for an interrupt signal when dump access
659 the remote tape through RSH. Thanks to Christian Weisgerber
660 <naddy@mips.rhein-neckar.de> for providing the patch.
662 5. Fixed a bug which was causing dump/restore to display
663 garbage characters instead of the remote host name.
665 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
666 =====================================================================
668 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
669 <kukuk@suse.de> for submitting the patch.
671 2. Fix for the exit code when using the size estimate option of
672 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
673 submitting the patch.
675 3. Handle EINTR in atomical reads/writes in dump, which was causing
676 dump to fail on some systems. Thanks to Eric Jergensen
677 <eric@dvns.com> for reporting the bug and submitting the patch.
679 4. Handle more than 16 characters for the device names in dumpdates.
680 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
681 tracking down the problem and proposing the solution.
683 5. Fixed a bug in dump which prevented the creation of the
684 dumpdates file when doing a 0-level dump without already
685 having a dumpdates file. Thanks to Patrik Schindler
686 <poc@pocnet.net> for reporting the bug.
688 6. Changed the way dump 'S' flag reports the size estimate
689 from number of blocks into bytes (making it compatible
690 with the Solaris version, and simplifying things for
691 amanda users). Thanks to Jason L Tibbitts III
692 <tibbs@math.uh.edu> for reporting the bug.
694 7. Fixed a compatibility problem in linux/alpha dump tape format.
695 Now the linux/alpha dump are (again) compatible with the
696 other dump formats. But this breaks compatibility with
697 older dumps made on alpha. Thanks to Mike Tibor
698 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
700 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
701 ======================================================================
703 1. Added a '--enable-kerberos' to configure.
705 2. Added a 'S' option to dump which determines the amount of space
706 that is needed to perform the dump without actually doing it, similar
707 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
708 <cermak@ahab.rutgers.edu>.
710 3. Added a 'M' multi-volume option to dump and restore which enables
711 dumping to multiple files (useful when dumping to an ext2
712 partition to make several dump files in order to bypass the 2GB
713 file size limitation). The 'f' argument is treated as a prefix and
714 the output files will be named <prefix>001, <prefix>002 etc. With
715 the 'M' flag, restore automatically selects the right file without
716 asking to enter a new tape each time.
718 4. Fixed a memory leak which was causing dump to grow very big
719 (270MB when dumping a 10GB filesystem...). Thanks to Jason
720 Fearon <jasonf@netrider.org.au> for reporting the bug.
722 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
723 ======================================================================
725 1. Make configure test if the system glob routines support
726 extended syntax (ALTDIRFUNC). If not, use the internal glob
727 routines instead of system ones. Thanks to Bernhard Sadlowski
728 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
729 and helping me resolve this and other minor libc5 compiling
732 2. Fix a problem when dumping a ext2fs with the 'filetype'
733 feature enabled. Thanks to Patrick J. LoPresti
734 <patl@cag.lcs.mit.edu> for reporting the bug and to
735 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
737 3. Made the nodump flag work on directories. A directory which
738 has the nodump flag gets never dumped, regardless of its
741 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
742 which allows dump on an active ext3 filesystem. However, this
743 is a "quick and dirty" patch which enables backup of an ext3
744 filesystem through the ext2 compatibility (by ignoring the
745 NEEDS_RECOVERY bit). The journal file is not recognized and
746 it is dumped (it should not).
748 5. Test the superblock compatibility flags when dumping, in order
749 to be sure that we know how to deal with specific features.
751 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
752 ====================================================================
754 1. Use lchown instead of chown, fixing a possible security problem
755 when restoring symlinks (a malicious user could use this
756 to deliberately corrupt the ownership of important system files).
757 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
758 this and providing the patch.
760 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
761 ====================================================================
763 1. Put dump sources under CVS, added Id tags in all files so
764 one can use 'ident' on binary files.
766 2. Added the dump/restore version in the usage text so one can
767 easily verify the version he is using.
769 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
770 a va_start/va_end problem on linux-ppc (always call va_start
771 va_end in pairs each time we use a vararg function).
773 4. Added again the DT_* constants because old libc does not
774 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
775 for submitting the bug report.
777 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
778 this should enable dumping big (huge) filesystems.
780 6. Added the RSH environment variable in order to be able to
781 use a rsh replacement like ssh when doing remote backups (and
782 bypass the security limitations of rcmd). Now you can do remote
783 backups without being root (or making dump setuid root).
785 7. Modified again the way dumpdates works. For incremental dumps,
786 we need to read dumpdates even if we are not using 'u' option.
787 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
790 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
791 ===================================================================
793 1. Removed the 'k' flag from the restore 'about' text if kerberos
796 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
797 to them (fsetflags takes a char*, setflags an open fd!).
799 3. (f)setflags is called only if the flags aren't empty. If the
800 file is a special file, a warning is printed, because changing
801 flags implies opening the device. Normally, a special file
802 should not have any flag... (Debian bug #29775, patch provided
803 by Abhijit Dasgupta <abhijit@ans.net>).
805 4. Made possible to dump a file system not mentioned in /etc/fstab.
806 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
808 5. Changed the default behaviour to not create dumpdates
809 unless 'u' option is specified. Removed the old "debian-patch"
810 which provided the same thing. (Debian bug #38136, #33818).
812 6. Removed all those dump*announce, since they were getting old...
814 7. Added warning messages if dumpdates does not exist and
815 when an inferior level dump does not exist (except for a level 0
818 8. Debugged the glob calls in interactive mode: restore used a
819 dirent struct which was different from the /usr/include/dirent.h
820 one (this used to work, is it a glibc2 change?), so none of the
821 compat glob (which used /usr/include/dirent.h) or the system glob
822 worked. Restore use now the system dirent (and the system
823 DT_* constants), which are compatible with BSD ones.
825 9. Added a configure flag (--with-dumpdatespath) to specify
826 the location of dumpdates file. By default, it is
829 10. Added the "AUTHOR" and "AVAILABILITY" sections and
830 included the current date/version in man pages.
832 11. Corrected the estimation of remaining time when
833 the operator doesn't change the tapes quickly enough. This
834 was an old bug, I thought I corrected it, and discovered
835 that in fact it was corrected in two different places, so
836 the results canceled each other...
838 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
839 ===================================================================
841 1. Integrated multiple patches from RedHat, Debian and SuSE:
843 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
845 - several fixes in the man pages.
846 - update the default tape device to /dev/st0.
847 - many updates for Linux Alpha (byte ordering, size_t etc).
849 - use environment variable for TMPDIR (instead of /tmp).
850 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
851 - workaround egcs bug (RedHat bugs #4281 and #2989).
852 - wire $(OPT) throughout Makefile's.
854 2. Upgrade the dump revision to 1, making possible to dump filesystems
855 made with e2fsprogs-1.15 or newer. Nothing seems to break...
857 3. Fix some compile warnings, prototype all functions.
859 4. Use glibc err/glob instead of internal compatibility
860 routines (only if available).
862 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
863 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
865 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
866 ======================================================================
868 1. Integrated the changes from FreeBSD-3.1-RELEASE
869 (mostly bug fixes, buffer overruns, dump has now an "automatic
870 tape length calculation" flag, dump/restore can use kerberos now
871 (this is NOT tested), use environment variables for TAPE and
874 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
876 3. Corrected a bug in restore when using 'C' option with multi-volumes
877 tapes (files splited accros two tapes give "size changed" errors
880 4. Corrected the long standing bug when dumping multiple tapes.
881 This works for me, needs further testing.
883 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
884 ====================================================================
886 1. Dump now runs correctly on kernels 2.1.x
887 Fix made by Gerald Peters <gapeters@worldnet.att.net>
889 Changes between versions 0.4b2 and 0.4b3
890 ========================================
892 1. Use realpath() if available
896 Changes between versions 0.4b1 and 0.4b2
897 ========================================
899 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
902 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
904 3. Fixed configure.in to correctly handle --enable-rmt
906 Changes between versions 0.3 and 0.4b1
907 ======================================
909 1. Integrated the changes from 4.4BSD-Lite2
911 2. Integrated the patches from Debian and Red Hat
913 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
915 4. Changed dump to use the Ext2fs library to get block addresses. This
916 should solve the endianness problem on SparcLinux.
918 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
919 distribution's one) to use autoconf
921 6. Fixed a few minor bugs
923 Changes between versions 0.2e and 0.2f
924 ======================================
926 1. Added the creation of named pipes (fifos) in restore.
928 2. Added the -N flag in the restore manual page.
930 3. Added the file kernel.patch which contains the llseek() optimization
931 patch for 1.2.x kernels.
933 4. Fixed a bug in the restoration of symbolic links: owner and group were
936 5. Integrated some changes from FreeBSD 2.2.
938 6. Added a call to ftruncate() after restoring each file to restore
939 correctly files ending by a hole.
941 Changes between versions 0.2d and 0.2e
942 ======================================
944 1. Fixed a bug in the "set owner/mode" process. Every file was restored
945 with owner = root (0) and group = root/wheel/whatever (0).
947 Changes between versions 0.2c and 0.2d
948 ======================================
950 1. Dump is now able to backup 2GB+ filesystems.
952 2. Dump and restore can now be linked as static binaries.
954 Changes between versions 0.2b and 0.2c
955 ======================================
957 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
959 Changes between versions 0.2a and 0.2b
960 ======================================
962 1. Really fixed the bug that I should have corrected in 0.2a.
964 2. Enabled optimization again.
966 Changes between versions 0.2 and 0.2a
967 =====================================
969 1. Disabled the optimization during compilation.
971 Changes between versions 0.1 and 0.2
972 ====================================
974 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
975 the fs_type field (actually, I modified the file fstab.c to make it
976 use the mntent functions).
978 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
979 made by dump should be readable by the BSD restore and backups made
980 by the BSD dump should be readable by restore. Unfortunately, this
981 means that the dump format has changed between version 0.1 and version
984 3. Dump is now able to backup a subtree, it is no longer limited to whole
985 filesystems like the BSD version.
987 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
990 Changes between versions 0.0 and 0.1
991 ====================================
993 1. Now create links rdump and rrestore during the `make install' step.
995 2. Linux port specific bugs added to the manual pages
997 3. Incorrect estimation of the number of tapes blocks fixed when doing
1000 4. Better ls-like format in restore in interactive mode.