1 $Id: CHANGES,v 1.158 2002/01/25 14:59:53 stelian Exp $
3 Changes between versions 0.4b26 and 0.4b27 (released ???????????????)
4 =====================================================================
6 1. Fixed behaviour of dump when exceeding resource limits
9 2. Added the -L flag to restore to allow the user to specify a
10 maximal allowed number of miscompares when using restore
11 with the -C option to check the backup.
13 3. Detailed the manual entry for the -N option of restore.
15 4. Added the -a flag to restore to make able doing unattended
16 restores in -i or -x mode (automatically walks through the
17 multiple dump volumes).
19 5. Extended the QFA mode to work with local files and/or
20 remote tapes and files. This way, restore can know in advance
21 the tape number and the offset for the inodes to extract and
22 can minimize the extraction time by seeking directly to the
25 6. Added the -A <archive> option to both dump and restore,
26 which makes dump to archive a dump table-of-contents in
27 the specified file to be used by restore to determine
28 whether a file is in the dump file that is being restored.
29 (the archive file syntax is also compatible with the
30 Solaris ufsdump generated one).
32 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
33 =====================================================================
35 1. Added a set of backup scripts from Eugenio Diaz
36 <getnito@yahoo.com> in the examples section. It features
37 automatic (cron based) full and incremental dumping of
38 several filesystems on a separate filesystem.
40 2. Fixed a off-by-one miscalculation which disabled dumping
41 a one letter subdirectory. Thanks to Chris Siebenmann
42 <cks@utcc.utoronto.ca> for reporting the bug.
44 3. Fixed several restore bugs occuring when trying to
45 restore some missing files on the tape. Thanks to Chris
46 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
48 4. Fixed --with-ldopts configure argument passing, installing from
49 a separate object directory, makefile cleanups contributed
50 by <splite@purdue.edu>.
52 5. Fix a bug which could caused, in some conditions, the highest
53 number inode of a filesystem, to not be dumped. Many thanks
54 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
57 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
58 =======================================================================
60 1. Added a mini howto from Patrick Walsh in the examples
63 2. Minor man pages syntax corrections. Thanks to
64 Chris Danis <screechco@home.com> for reporting the bugs.
66 3. Added a script from David B. Peterson <dave@toppledwagon.com>
67 to the examples section. It features dumping several
68 filesystems to a remote tape drive upon ssh.
70 4. Added a patch provided by Richard Jones <rich@annexia.org>
71 which allows BRADEMAX (number of read errors tolerated by
72 dump) to be adjusted using the -I option.
74 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
75 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
78 6. Add the -l option to restore to specify if, when doing a
79 remote restore, the file used to access the backup is a
80 regular file (the defaults being a tape drive). Restore needs
81 to know this information when reading a remote compressed
82 dump. Previously, this information was autodetected, but
83 the autodetection code fails (with ioctl: Inappropriate ioctl
84 for device) when using a non Linux remote box. Thanks to
85 many users and especially to Eros Albertazzi
86 <eros@lamel.bo.cnr.it> for reporting this.
88 7. Found a workaroung for the dump deadlock problem (3 childs
89 stuck in pause(), father in read()). The workaround seems
90 to work for me and several beta-testers. If it doesn't work
91 for you, please report back.
93 8. Updated the RPM spec file (BuildPrereq, URL etc).
95 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
96 ========================================================================
98 1. Fixed the permissions of a newly created QFA file by dump.
100 2. Cleaned up the source of dump (the external variables
101 definition was a complete mess, making possible to have
104 3. Fixed restore to use the full tape volume path when doing
105 a compare (since it changes the working directory to the
106 filesystem being compared in the process).
108 4. Added the -q option to dump which makes dump abort
109 whenever it needs operator attention. It should be
110 easier to use dump in scripts with this option.
112 5. Detect the use of incompatible options to dump and
113 refuse them (like -a and -B options together).
115 6. Added bzip2 compression to dump/restore (use option -j level
116 to select it). Note that you will need a restore version
117 >= 0.4b24 in order to restore a bzip2 compressed dump.
118 The same warning as for the zlib compression applies:
119 the tape format of a bzip2 dump is not compatible with the
120 original BSD tape format.
122 7. Fixed a overflow problem in dump corrupting the dump when
123 very large files were encountered. Thanks to Vladimir Nadvornik
126 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
127 kernel buffer/page cache before starting the dump, helping
128 a bit those who use dump on mounted filesystems. Thanks to
129 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
132 9. Updated the RPM spec file following the RedHat changes
133 (dynamically linked binaries now in /usr/sbin etc).
135 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
136 which enables restore to recognise multi volume compressed dumps
137 done on CD. Included his scripts for dump (which pipe the dump
138 output directly into cdrecord) and restore. There is now
139 possible to to backups to CD on the fly!
141 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
142 ===================================================================
144 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
145 Marcus Meissner (Caldera International Security Dept.).
147 2. Implement the Sun rmt extended protocol. Patch submitted
148 by Ian Gordon <iangordon@users.sourceforge.net>.
150 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
151 the linux kernel header. This ensures that dump always has the
152 latest version of this file. Patch submitted by
153 Andreas Dilger <adilger@turbolinux.com>.
155 4. Report any filesystem present in either /etc/fstab with a
156 positive passno or /etc/dumpdates in dump -w output.
157 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
159 5. Fixed the looping problem in dump introduced in the
162 6. Changed the -B option of dump to limit the size of
163 _compressed_ output per volume if compression is on.
164 Patch contributed by Helmut Jarausch
165 <jarausch@igpm.rwth-aachen.de>. Note however that, since
166 it is impossible to predict the size of the compressed
167 data before doing the compression, the -B limit is a bit
170 7. Fixed a bug in reading the operator typed file/tape path for
171 the next volume in restore.
173 8. Implemented a "-F script" option to restore which permits the
174 user to specify a script which will be launched at the
175 beginning of each tape, useful for automatic programming of
176 tape changers for example. See the restore man page for the
177 script parameters and return codes.
179 9. Small fix for the QFA routines provided by Uwe Gohlke
180 <uwe@ugsoft.de>, and some recommendations for QFA uses in
183 10. Fixed the multivolume restoring where making a mistake
184 to the 'Mount next tape' prompt caused several blocks to
187 11. Enhanced the -e option of dump to take as a parameter a
188 comma separated list of inode numbers.
190 12. Added the -E option to dump which specify a file containing
191 inode numbers to exclude from the dump.
193 13. Fixed the compressed multi-volume dump + restore.
195 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
196 ==================================================================
198 1. Made dump report the number of blocks dumped per volume.
199 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
201 2. Fix a bug in dump where waiting too much at the 'change volumes'
202 question caused the volume to be rewritten. Thanks to
203 Graham Reed <greed@users.sourceforge.net> for reporting the
204 bug and providing a patch.
206 3. Added a compression option to dump, contributed by Jerry
207 Peters <gapeters@worldnet.att.net>.
209 WARNING: the compression works only on files/pipes and
210 drives supporting variable block size.
212 WARNING2: tapes created using the compression feature are
213 incompatible with the standard dump tape format, and a
214 version of restore >= 0.4b22 will be needed for extraction.
216 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
217 architectures. Thanks to Paul Slootman <paul@debian.org> for
218 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
221 5. Many cleanups (CPP defines, const char warnings, check of
222 ext2fs COMPAT flags, time_t cleanups, added libext2 version
223 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
225 6. Made --prefix option work in configure. All the install path
226 are now based on the configure parameters.
228 7. Added the Quick File Access mode in dump/restore, contributed
229 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
230 a file tape position for each inode, and this file is used by
231 restore (if called with parameter Q and the filename)
232 to directly position the tape at the file restore is currently
233 working on. This saves hours when restoring single files from
234 large backups, saves the tapes and the drive's head. Use
235 --enable-qfa option of configure to compile in the QFA support.
237 8. Added the possibility to dump several files and directories
238 in a single invocation of dump. Thanks to Uwe Gohlke
239 <uwe@ugsoft.de> for implementing this option.
241 9. Fixed the dumping and restoring of files with holes
242 (files ending with holes didn't get dumped or restored
245 10. Fixed a socket descriptor leak in dump, which leaved opened
246 3 file descriptors per dump process (and there is one dump
249 11. Fixed dump large file system support, by explicit use of
250 open64/lseek64/etc functions (explicit use needed because
251 e2fsprogs libraries don't behave well when compiled with
252 FILE_OFFSET_BITS=64).
254 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
255 ======================================================================
257 1. Fixed some bugs in the dump -w|-W logic introduced by
258 the previous version. Thanks to Andreas Dilger
259 <adilger@turbolinux.com> for his help on this one.
261 2. Fixed again a compilation problem when using old e2fs
262 headers (filesystem label related). Thanks to many users
263 who reported this stupid error.
265 3. Fixed a build problem on old lib5 based systems dealing with
266 _PATH_MNTTAB being undefined. Thanks to John Adams
267 <johna@onevista.com> for reporting the bug.
269 4. Improved the error detection in restore when creating
270 the temporary files in TMPDIR. Now restore will corectly
271 report a 'no space left on device' error instead of
272 strange errors which could imply an archive corruption.
273 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
274 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
276 5. Added the throughput information in dump's progression
277 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
280 6. Use libext2fs's inode scanning routines, which are particularly
281 robust when dealing with errors (especially when having some
282 bad blocks on the filesystem). Many thanks to Colin
283 <colin@horizon.com> for reporting the bug, investigating
284 the issues, finding a workaround, writing the patch and
285 fully testing it... (of course, if this patch will break
286 anything, he is to blame for :-)).
288 7. Made dump and restore LFS aware. Dump can dump a filesystem
289 containing large files, generate a large file on output and
290 restore can restore them. This feature is not enabled by
291 default, use --enable-largefile option of configure to enable
292 it (you need to have a LFS aware glibc though). Thanks to
293 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
294 and to Theodore T'so <tytso@valinux.com> for his always
297 8. Made dump ask upon a tape write error if it should rewrite
298 the current volume (assume this is a bad tape) or if it should
299 assume an end-of-tape condition (useful for tape drives which
300 behaves badly at the end of the tape). Thanks to Andreas
301 Dilger <adilger@turbolinux.com> for the suggestion.
303 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
304 =======================================================================
306 1. Fixed a small compilation problem due to a change
307 in the definintion of the struct sigaction in
308 glibc 2.0 / libc5. Thanks to Gunther Schlegel
309 <schlegel@riege.de> for reporting the bug and to
310 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
313 2. Modified the label and uuid analysis in order to be
314 self-contained (independant of kernel/libc headers). This
315 should fix the compile with older kernel/libc headers and
316 will preserve the functionality. Thanks to Bernhard Erdmann
317 <bernhard.erdmann@gmx.de> for reporting the bug.
319 3. The 'exclude inode' option, if applied to a directory
320 inode, excludes now correctly all the leaves of this
321 directory. Thanks to John R. Dennison
322 <gerdesas@users.sourceforge.net> for reporting the bug.
324 4. Fixed the '-e' option to disable the possibility
325 to exclude the root inode (which causes the dumps to
326 be unreadable by restore). Prevented array overflow
327 when multiple -e options are used.
329 5. Fixed dump to correctly interpret a filesystem argument
330 which is a mountpoint and it is not an absolute path
331 (as specified in the fstab). Thanks to Bernhard R. Erdmann
332 <be@berdmann.de> for reporting the bug.
334 6. Made dump able to backup files larger than 2 GB. Note that
335 dump still doesn't cope with files larger than 4 GB.
337 7. Restore the real uid and gid before invoking an external
338 RSH program (big hole when dump or restore is suid root!).
340 8. Get the values for the default tape device and the location
341 of fstab file from the system headers. Thanks to
342 Andreas Dilger <adilger@turbolinux.com> for the patch.
344 9. Made dump -w|-W report all recognized filesystems
345 present in either /etc/fstab or /etc/dumpdates, and present
346 the list in the same order as in fstab file. Thanks
347 to Andreas Dilger <adilger@turbolinux.com> for the patch.
349 10. Made dump's -a (automatic end of tape detection) the
350 default. Specifying one of -B, -c, -d or -s options will
351 override the EOM detection. Thanks to Andreas Dilger
352 <adilger@turbolinux.com> for the patch.
354 11. Save the ext2 filesystem label into the dump volume label.
355 Specifying a volume label on the command line (-L option)
356 will override this feature. Thanks to Andreas Dilger
357 <adilger@turbolinux.com> for the patch.
359 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
360 =====================================================================
362 1. Fixed the signal handling in dump (which I broke in 0.4b17)
363 which was causing several strange dump failures (dump
364 hanged or segmentation faults etc).
366 2. Specified the default blocksize in the dump man page.
368 3. Changed two info messages of restore to be written on stdout
369 instead of stderr in order to leave stderr only for errors
370 or warnings. Thanks to Stephen Atwell
371 <satwell@urbana.css.mot.com> for the suggestion.
373 4. Corrected an off by one calculation which prevented
374 dumping a directory having a 1 character name length.
375 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
376 for reporting the bug.
378 5. Reinforce dump in order to allow only level 0 and no
379 -u option when dumping a subdirectory, and document
380 this in the man page. Thanks to Bernhard Erdmann
381 <bernhard.erdmann@gmx.de> for reporting the bug.
383 6. Fixed a small harmless bug in restore which caused
384 in some conditions hard links to be restored several
385 times (therefore generation some warning messages).
386 Thanks to Kenneth Porter <shiva@well.com> for
389 7. Updated the RPM spec file to the latest RedHat version,
390 providing FHS packaging and other cosmetic changes.
391 You will need RPM version >= 3.0.5 in order to build the RPM.
393 8. Updated the configure script to check for libtermcap
394 before checking for libreadline, because we need this
395 library in order to compile the readline support.
397 9. Made dump understand the LABEL= and UUID= notation
398 both in /etc/fstab and on the command line. Note that
399 you will need the /proc filesystem in order to use
400 these notations. Thanks to Erik Troan <ewt@redhat.com>
401 for providing the patch.
403 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
404 ===================================================================
406 1. Fixed a potential buffer overflow in restore. Thanks
407 to Stan Bubrouski <satan@fastdial.net> for reporting
410 2. Fixed a readline-related bug which prevented
411 'cat DUMPFILE | restore -i -f -' from working. Thanks
412 to Charles Karney <karney@users.sourceforge.net>
415 3. Changed a few "panic" into "exit", causing restore to
416 be more stable against some attacks (like the last one
417 reported on Bugtraq, although the last version of restore
418 was not vulnerable - just dumped core). Thanks to
419 Andreas Hasenack <andreas@conectiva.com.br> for reporting
422 4. Removed the suid-root bit on dump and restore in the
423 default build (and generated RPMs). It should be safer
424 now. Users who need the suid-root capabilities in order
425 to make network backups should read first the man page
426 of dump and enable the suid bit manually.
428 5. Added -ltermcap to the compile parameters for restore
429 when using readline support, in order to make the compile
430 process work on some systems (SuSE ?). Thanks to
431 Patrik Schindler <poc@pocnet.net> for reporting the bug.
433 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
434 ====================================================================
436 1. The -F script is called now *only* at the end of a tape,
437 not at the end of the dump. Thanks to Christian Haul
438 <haul@informatik.tu-darmstadt.de> for the bug report.
440 Normally, the device name and the current volume number
441 are passed on the command line of the script. However,
442 if you want the old style script (with no arguments
443 passed on the command line), you can enable it in
444 configure with the --enable-oldstylefscript.
446 2. Use posix signal handling to preserve dump functionality
447 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
450 3. Made the exit code of restore in case of a 'C'ompare
451 command reflect the result of the compare. An exit status
452 of 0 means the dump archive is correct, 1 means tape errors,
453 2 means that some files were modified. Thanks to Kenneth Porter
454 <shiva@well.com> for the suggestion.
456 4. Made (finally) quotes work in interactive restore.
458 5. Small fixes in order to allow dump to compile with a
459 really old e2fsprogs version. Thanks to Ian Zimmerman
460 <itz@speakeasy.org> for the bug report.
462 6. Add GNU readline capabilities to interactive restore.
463 Use configure's --enable-readline flag to enable this feature.
464 Thanks to Patrik Schindler <poc@pocnet.net> for the
467 7. Do the compare on the fly in restore 'C' mode (this will
468 allow not to exhaust the available /tmp space when
469 ccmparing large files). Thanks to Kenneth Porter
470 <shiva@well.com> for the suggestion.
472 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
473 ====================================================================
475 1. Fixed some several duplicate 'DUMP: DUMP:' in the
478 2. Corrected the estimation of blocks to dump. Note that
479 this is still wrong for large directory inodes, due
480 to the size of a BSD directory entry on the tape
481 (estimation is based on the size of the ext2 directory,
482 which is smaller because it doesn't contain the null
483 character at the end).
485 3. Made dump report the total number of MB written to
486 tape at the end of run. Thanks to W. Reilly Cooley
487 <wcooley@nakedape.cc> for the patch.
489 4. Added the -X option to restore giving the possibility
490 to read the names of the files to be extracted/listed
491 from a text file (in addition of the command line).
492 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
495 5. Added the device name and the current volume number
496 as arguments to the end of tape script (dump -F option).
498 6. Made the multi-volume dump work again (after having
499 broken it in 0.4b15).
501 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
502 ===================================================================
504 1. Added a prompt command in interactive restore mode. Thanks
505 to Andreas Dilger <adilger@home.com> for the patch.
507 2. Fixed a buffer overflow problem in dump (caused by
508 not checking the size of the filesystem parameter).
509 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
510 reporting this on Bugtraq (and to several dump users
511 who forwarded me his mail).
513 3. Added the '-F script' option to dump in order to
514 launch a script at the end of each tape (to be used
515 with a tape changer, or to notify the sysadmin by
518 4. Fixed a bug in restore compare code caused by the changes
521 5. Fixed the treatment of options using the old BSD syntax
522 in both dump and restore.
524 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
525 =======================================================================
527 1. Fixed a bug in dump which may have caused invalid deleted
528 directories to be dumped out if they were deleted after the
529 mapping pass. This could occure on active filesystem and lead
530 to heap corruption (causing dump malfunction in many possible ways).
531 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
532 bug and submitting the patch.
534 2. Fixed the handling of the filesystem parameter in dump. A
535 '/mnt/disk' parameter caused the disk contents to be dumped,
536 but a '/mnt/disk/' parameter caused the mountpoint directory
537 to be dumped (generally an empty directory).
539 3. Improved the output of dump in order to tell which directory
540 it is currently dumping (when dumping a subtree).
542 4. Added the '-e' exclude inode option to dump. Thanks to
543 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
545 5. Added a REPORTING-BUGS file in order to provide a guide
546 on how to correctly report a bug in dump/restore.
548 6. Fixed a restore problem when restoring a hard link to an inode
549 having the immutable or append-only attribute set. Thanks to
550 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
552 7. Fixed a compatibility problem between dumps made on little
553 endian machines (the format was correct) and big endian
554 machines (the format was incorrect). This fix break the
555 compatibility with the older dumps made on big endian
556 machines (sparc, m86k, ppc etc). For the first time in
557 linux dump's history (I believe), the dumps made by *BSD,
558 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
559 so interchangeable. Thanks to Rob Cermak
560 <cermak@ahab.rutgers.edu> for submitting the bug and
561 helping me test the fix.
563 8. Fixed the way dump reports the remaining percent/time, if
564 the number of blocks actually dumped exceeds the estimated
565 number of blocks. Thanks to Jean-Paul van der Jagt
566 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
568 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
569 ======================================================================
571 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
572 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
574 2. Removed the requirement to build the RPM as root from the
575 spec file. Thanks to Christian Weisgerber
576 <naddy@mips.rhein-neckar.de> for submitting this.
578 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
579 causing dump to block when dumping really huge filesystems.
580 Many thanks to Patrik Schindler <poc@pocnet.net> for
581 helping me find this bug.
583 4. Fixed the treatment for an interrupt signal when dump access
584 the remote tape through RSH. Thanks to Christian Weisgerber
585 <naddy@mips.rhein-neckar.de> for providing the patch.
587 5. Fixed a bug which was causing dump/restore to display
588 garbage characters instead of the remote host name.
590 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
591 =====================================================================
593 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
594 <kukuk@suse.de> for submitting the patch.
596 2. Fix for the exit code when using the size estimate option of
597 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
598 submitting the patch.
600 3. Handle EINTR in atomical reads/writes in dump, which was causing
601 dump to fail on some systems. Thanks to Eric Jergensen
602 <eric@dvns.com> for reporting the bug and submitting the patch.
604 4. Handle more than 16 characters for the device names in dumpdates.
605 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
606 tracking down the problem and proposing the solution.
608 5. Fixed a bug in dump which prevented the creation of the
609 dumpdates file when doing a 0-level dump without already
610 having a dumpdates file. Thanks to Patrik Schindler
611 <poc@pocnet.net> for reporting the bug.
613 6. Changed the way dump 'S' flag reports the size estimate
614 from number of blocks into bytes (making it compatible
615 with the Solaris version, and simplifying things for
616 amanda users). Thanks to Jason L Tibbitts III
617 <tibbs@math.uh.edu> for reporting the bug.
619 7. Fixed a compatibility problem in linux/alpha dump tape format.
620 Now the linux/alpha dump are (again) compatible with the
621 other dump formats. But this breaks compatibility with
622 older dumps made on alpha. Thanks to Mike Tibor
623 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
625 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
626 ======================================================================
628 1. Added a '--enable-kerberos' to configure.
630 2. Added a 'S' option to dump which determines the amount of space
631 that is needed to perform the dump without actually doing it, similar
632 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
633 <cermak@ahab.rutgers.edu>.
635 3. Added a 'M' multi-volume option to dump and restore which enables
636 dumping to multiple files (useful when dumping to an ext2
637 partition to make several dump files in order to bypass the 2GB
638 file size limitation). The 'f' argument is treated as a prefix and
639 the output files will be named <prefix>001, <prefix>002 etc. With
640 the 'M' flag, restore automatically selects the right file without
641 asking to enter a new tape each time.
643 4. Fixed a memory leak which was causing dump to grow very big
644 (270MB when dumping a 10GB filesystem...). Thanks to Jason
645 Fearon <jasonf@netrider.org.au> for reporting the bug.
647 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
648 ======================================================================
650 1. Make configure test if the system glob routines support
651 extended syntax (ALTDIRFUNC). If not, use the internal glob
652 routines instead of system ones. Thanks to Bernhard Sadlowski
653 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
654 and helping me resolve this and other minor libc5 compiling
657 2. Fix a problem when dumping a ext2fs with the 'filetype'
658 feature enabled. Thanks to Patrick J. LoPresti
659 <patl@cag.lcs.mit.edu> for reporting the bug and to
660 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
662 3. Made the nodump flag work on directories. A directory which
663 has the nodump flag gets never dumped, regardless of its
666 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
667 which allows dump on an active ext3 filesystem. However, this
668 is a "quick and dirty" patch which enables backup of an ext3
669 filesystem through the ext2 compatibility (by ignoring the
670 NEEDS_RECOVERY bit). The journal file is not recognized and
671 it is dumped (it should not).
673 5. Test the superblock compatibility flags when dumping, in order
674 to be sure that we know how to deal with specific features.
676 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
677 ====================================================================
679 1. Use lchown instead of chown, fixing a possible security problem
680 when restoring symlinks (a malicious user could use this
681 to deliberately corrupt the ownership of important system files).
682 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
683 this and providing the patch.
685 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
686 ====================================================================
688 1. Put dump sources under CVS, added Id tags in all files so
689 one can use 'ident' on binary files.
691 2. Added the dump/restore version in the usage text so one can
692 easily verify the version he is using.
694 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
695 a va_start/va_end problem on linux-ppc (always call va_start
696 va_end in pairs each time we use a vararg function).
698 4. Added again the DT_* constants because old libc does not
699 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
700 for submitting the bug report.
702 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
703 this should enable dumping big (huge) filesystems.
705 6. Added the RSH environment variable in order to be able to
706 use a rsh replacement like ssh when doing remote backups (and
707 bypass the security limitations of rcmd). Now you can do remote
708 backups without being root (or making dump setuid root).
710 7. Modified again the way dumpdates works. For incremental dumps,
711 we need to read dumpdates even if we are not using 'u' option.
712 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
715 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
716 ===================================================================
718 1. Removed the 'k' flag from the restore 'about' text if kerberos
721 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
722 to them (fsetflags takes a char*, setflags an open fd!).
724 3. (f)setflags is called only if the flags aren't empty. If the
725 file is a special file, a warning is printed, because changing
726 flags implies opening the device. Normally, a special file
727 should not have any flag... (Debian bug #29775, patch provided
728 by Abhijit Dasgupta <abhijit@ans.net>).
730 4. Made possible to dump a file system not mentioned in /etc/fstab.
731 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
733 5. Changed the default behaviour to not create dumpdates
734 unless 'u' option is specified. Removed the old "debian-patch"
735 which provided the same thing. (Debian bug #38136, #33818).
737 6. Removed all those dump*announce, since they were getting old...
739 7. Added warning messages if dumpdates does not exist and
740 when an inferior level dump does not exist (except for a level 0
743 8. Debugged the glob calls in interactive mode: restore used a
744 dirent struct which was different from the /usr/include/dirent.h
745 one (this used to work, is it a glibc2 change?), so none of the
746 compat glob (which used /usr/include/dirent.h) or the system glob
747 worked. Restore use now the system dirent (and the system
748 DT_* constants), which are compatible with BSD ones.
750 9. Added a configure flag (--with-dumpdatespath) to specify
751 the location of dumpdates file. By default, it is
754 10. Added the "AUTHOR" and "AVAILABILITY" sections and
755 included the current date/version in man pages.
757 11. Corrected the estimation of remaining time when
758 the operator doesn't change the tapes quickly enough. This
759 was an old bug, I thought I corrected it, and discovered
760 that in fact it was corrected in two different places, so
761 the results canceled each other...
763 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
764 ===================================================================
766 1. Integrated multiple patches from RedHat, Debian and SuSE:
768 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
770 - several fixes in the man pages.
771 - update the default tape device to /dev/st0.
772 - many updates for Linux Alpha (byte ordering, size_t etc).
774 - use environment variable for TMPDIR (instead of /tmp).
775 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
776 - workaround egcs bug (RedHat bugs #4281 and #2989).
777 - wire $(OPT) throughout Makefile's.
779 2. Upgrade the dump revision to 1, making possible to dump filesystems
780 made with e2fsprogs-1.15 or newer. Nothing seems to break...
782 3. Fix some compile warnings, prototype all functions.
784 4. Use glibc err/glob instead of internal compatibility
785 routines (only if available).
787 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
788 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
790 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
791 ======================================================================
793 1. Integrated the changes from FreeBSD-3.1-RELEASE
794 (mostly bug fixes, buffer overruns, dump has now an "automatic
795 tape length calculation" flag, dump/restore can use kerberos now
796 (this is NOT tested), use environment variables for TAPE and
799 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
801 3. Corrected a bug in restore when using 'C' option with multi-volumes
802 tapes (files splited accros two tapes give "size changed" errors
805 4. Corrected the long standing bug when dumping multiple tapes.
806 This works for me, needs further testing.
808 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
809 ====================================================================
811 1. Dump now runs correctly on kernels 2.1.x
812 Fix made by Gerald Peters <gapeters@worldnet.att.net>
814 Changes between versions 0.4b2 and 0.4b3
815 ========================================
817 1. Use realpath() if available
821 Changes between versions 0.4b1 and 0.4b2
822 ========================================
824 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
827 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
829 3. Fixed configure.in to correctly handle --enable-rmt
831 Changes between versions 0.3 and 0.4b1
832 ======================================
834 1. Integrated the changes from 4.4BSD-Lite2
836 2. Integrated the patches from Debian and Red Hat
838 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
840 4. Changed dump to use the Ext2fs library to get block addresses. This
841 should solve the endianness problem on SparcLinux.
843 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
844 distribution's one) to use autoconf
846 6. Fixed a few minor bugs
848 Changes between versions 0.2e and 0.2f
849 ======================================
851 1. Added the creation of named pipes (fifos) in restore.
853 2. Added the -N flag in the restore manual page.
855 3. Added the file kernel.patch which contains the llseek() optimization
856 patch for 1.2.x kernels.
858 4. Fixed a bug in the restoration of symbolic links: owner and group were
861 5. Integrated some changes from FreeBSD 2.2.
863 6. Added a call to ftruncate() after restoring each file to restore
864 correctly files ending by a hole.
866 Changes between versions 0.2d and 0.2e
867 ======================================
869 1. Fixed a bug in the "set owner/mode" process. Every file was restored
870 with owner = root (0) and group = root/wheel/whatever (0).
872 Changes between versions 0.2c and 0.2d
873 ======================================
875 1. Dump is now able to backup 2GB+ filesystems.
877 2. Dump and restore can now be linked as static binaries.
879 Changes between versions 0.2b and 0.2c
880 ======================================
882 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
884 Changes between versions 0.2a and 0.2b
885 ======================================
887 1. Really fixed the bug that I should have corrected in 0.2a.
889 2. Enabled optimization again.
891 Changes between versions 0.2 and 0.2a
892 =====================================
894 1. Disabled the optimization during compilation.
896 Changes between versions 0.1 and 0.2
897 ====================================
899 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
900 the fs_type field (actually, I modified the file fstab.c to make it
901 use the mntent functions).
903 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
904 made by dump should be readable by the BSD restore and backups made
905 by the BSD dump should be readable by restore. Unfortunately, this
906 means that the dump format has changed between version 0.1 and version
909 3. Dump is now able to backup a subtree, it is no longer limited to whole
910 filesystems like the BSD version.
912 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
915 Changes between versions 0.0 and 0.1
916 ====================================
918 1. Now create links rdump and rrestore during the `make install' step.
920 2. Linux port specific bugs added to the manual pages
922 3. Incorrect estimation of the number of tapes blocks fixed when doing
925 4. Better ls-like format in restore in interactive mode.