1 $Id: CHANGES,v 1.150 2001/12/24 12:31:12 stelian Exp $
3 Changes between versions 0.4b25 and 0.4b26 (released ?????????????????)
4 =======================================================================
6 1. Added a set of backup scripts from Eugenio Diaz
7 <getnito@yahoo.com> in the examples section. It features
8 automatic (cron based) full and incremental dumping of
9 several filesystems on a separate filesystem.
11 2. Fixed a off-by-one miscalculation which disabled dumping
12 a one letter subdirectory. Thanks to Chris Siebenmann
13 <cks@utcc.utoronto.ca> for reporting the bug.
15 3. Fixed several restore bugs occuring when trying to
16 restore some missing files on the tape. Thanks to Chris
17 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
19 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
20 =======================================================================
22 1. Added a mini howto from Patrick Walsh in the examples
25 2. Minor man pages syntax corrections. Thanks to
26 Chris Danis <screechco@home.com> for reporting the bugs.
28 3. Added a script from David B. Peterson <dave@toppledwagon.com>
29 to the examples section. It features dumping several
30 filesystems to a remote tape drive upon ssh.
32 4. Added a patch provided by Richard Jones <rich@annexia.org>
33 which allows BRADEMAX (number of read errors tolerated by
34 dump) to be adjusted using the -I option.
36 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
37 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
40 6. Add the -l option to restore to specify if, when doing a
41 remote restore, the file used to access the backup is a
42 regular file (the defaults being a tape drive). Restore needs
43 to know this information when reading a remote compressed
44 dump. Previously, this information was autodetected, but
45 the autodetection code fails (with ioctl: Inappropriate ioctl
46 for device) when using a non Linux remote box. Thanks to
47 many users and especially to Eros Albertazzi
48 <eros@lamel.bo.cnr.it> for reporting this.
50 7. Found a workaroung for the dump deadlock problem (3 childs
51 stuck in pause(), father in read()). The workaround seems
52 to work for me and several beta-testers. If it doesn't work
53 for you, please report back.
55 8. Updated the RPM spec file (BuildPrereq, URL etc).
57 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
58 ========================================================================
60 1. Fixed the permissions of a newly created QFA file by dump.
62 2. Cleaned up the source of dump (the external variables
63 definition was a complete mess, making possible to have
66 3. Fixed restore to use the full tape volume path when doing
67 a compare (since it changes the working directory to the
68 filesystem being compared in the process).
70 4. Added the -q option to dump which makes dump abort
71 whenever it needs operator attention. It should be
72 easier to use dump in scripts with this option.
74 5. Detect the use of incompatible options to dump and
75 refuse them (like -a and -B options together).
77 6. Added bzip2 compression to dump/restore (use option -j level
78 to select it). Note that you will need a restore version
79 >= 0.4b24 in order to restore a bzip2 compressed dump.
80 The same warning as for the zlib compression applies:
81 the tape format of a bzip2 dump is not compatible with the
82 original BSD tape format.
84 7. Fixed a overflow problem in dump corrupting the dump when
85 very large files were encountered. Thanks to Vladimir Nadvornik
88 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
89 kernel buffer/page cache before starting the dump, helping
90 a bit those who use dump on mounted filesystems. Thanks to
91 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
94 9. Updated the RPM spec file following the RedHat changes
95 (dynamically linked binaries now in /usr/sbin etc).
97 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
98 which enables restore to recognise multi volume compressed dumps
99 done on CD. Included his scripts for dump (which pipe the dump
100 output directly into cdrecord) and restore. There is now
101 possible to to backups to CD on the fly!
103 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
104 ===================================================================
106 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
107 Marcus Meissner (Caldera International Security Dept.).
109 2. Implement the Sun rmt extended protocol. Patch submitted
110 by Ian Gordon <iangordon@users.sourceforge.net>.
112 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
113 the linux kernel header. This ensures that dump always has the
114 latest version of this file. Patch submitted by
115 Andreas Dilger <adilger@turbolinux.com>.
117 4. Report any filesystem present in either /etc/fstab with a
118 positive passno or /etc/dumpdates in dump -w output.
119 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
121 5. Fixed the looping problem in dump introduced in the
124 6. Changed the -B option of dump to limit the size of
125 _compressed_ output per volume if compression is on.
126 Patch contributed by Helmut Jarausch
127 <jarausch@igpm.rwth-aachen.de>. Note however that, since
128 it is impossible to predict the size of the compressed
129 data before doing the compression, the -B limit is a bit
132 7. Fixed a bug in reading the operator typed file/tape path for
133 the next volume in restore.
135 8. Implemented a "-F script" option to restore which permits the
136 user to specify a script which will be launched at the
137 beginning of each tape, useful for automatic programming of
138 tape changers for example. See the restore man page for the
139 script parameters and return codes.
141 9. Small fix for the QFA routines provided by Uwe Gohlke
142 <uwe@ugsoft.de>, and some recommendations for QFA uses in
145 10. Fixed the multivolume restoring where making a mistake
146 to the 'Mount next tape' prompt caused several blocks to
149 11. Enhanced the -e option of dump to take as a parameter a
150 comma separated list of inode numbers.
152 12. Added the -E option to dump which specify a file containing
153 inode numbers to exclude from the dump.
155 13. Fixed the compressed multi-volume dump + restore.
157 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
158 ==================================================================
160 1. Made dump report the number of blocks dumped per volume.
161 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
163 2. Fix a bug in dump where waiting too much at the 'change volumes'
164 question caused the volume to be rewritten. Thanks to
165 Graham Reed <greed@users.sourceforge.net> for reporting the
166 bug and providing a patch.
168 3. Added a compression option to dump, contributed by Jerry
169 Peters <gapeters@worldnet.att.net>.
171 WARNING: the compression works only on files/pipes and
172 drives supporting variable block size.
174 WARNING2: tapes created using the compression feature are
175 incompatible with the standard dump tape format, and a
176 version of restore >= 0.4b22 will be needed for extraction.
178 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
179 architectures. Thanks to Paul Slootman <paul@debian.org> for
180 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
183 5. Many cleanups (CPP defines, const char warnings, check of
184 ext2fs COMPAT flags, time_t cleanups, added libext2 version
185 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
187 6. Made --prefix option work in configure. All the install path
188 are now based on the configure parameters.
190 7. Added the Quick File Access mode in dump/restore, contributed
191 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
192 a file tape position for each inode, and this file is used by
193 restore (if called with parameter Q and the filename)
194 to directly position the tape at the file restore is currently
195 working on. This saves hours when restoring single files from
196 large backups, saves the tapes and the drive's head. Use
197 --enable-qfa option of configure to compile in the QFA support.
199 8. Added the possibility to dump several files and directories
200 in a single invocation of dump. Thanks to Uwe Gohlke
201 <uwe@ugsoft.de> for implementing this option.
203 9. Fixed the dumping and restoring of files with holes
204 (files ending with holes didn't get dumped or restored
207 10. Fixed a socket descriptor leak in dump, which leaved opened
208 3 file descriptors per dump process (and there is one dump
211 11. Fixed dump large file system support, by explicit use of
212 open64/lseek64/etc functions (explicit use needed because
213 e2fsprogs libraries don't behave well when compiled with
214 FILE_OFFSET_BITS=64).
216 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
217 ======================================================================
219 1. Fixed some bugs in the dump -w|-W logic introduced by
220 the previous version. Thanks to Andreas Dilger
221 <adilger@turbolinux.com> for his help on this one.
223 2. Fixed again a compilation problem when using old e2fs
224 headers (filesystem label related). Thanks to many users
225 who reported this stupid error.
227 3. Fixed a build problem on old lib5 based systems dealing with
228 _PATH_MNTTAB being undefined. Thanks to John Adams
229 <johna@onevista.com> for reporting the bug.
231 4. Improved the error detection in restore when creating
232 the temporary files in TMPDIR. Now restore will corectly
233 report a 'no space left on device' error instead of
234 strange errors which could imply an archive corruption.
235 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
236 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
238 5. Added the throughput information in dump's progression
239 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
242 6. Use libext2fs's inode scanning routines, which are particularly
243 robust when dealing with errors (especially when having some
244 bad blocks on the filesystem). Many thanks to Colin
245 <colin@horizon.com> for reporting the bug, investigating
246 the issues, finding a workaround, writing the patch and
247 fully testing it... (of course, if this patch will break
248 anything, he is to blame for :-)).
250 7. Made dump and restore LFS aware. Dump can dump a filesystem
251 containing large files, generate a large file on output and
252 restore can restore them. This feature is not enabled by
253 default, use --enable-largefile option of configure to enable
254 it (you need to have a LFS aware glibc though). Thanks to
255 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
256 and to Theodore T'so <tytso@valinux.com> for his always
259 8. Made dump ask upon a tape write error if it should rewrite
260 the current volume (assume this is a bad tape) or if it should
261 assume an end-of-tape condition (useful for tape drives which
262 behaves badly at the end of the tape). Thanks to Andreas
263 Dilger <adilger@turbolinux.com> for the suggestion.
265 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
266 =======================================================================
268 1. Fixed a small compilation problem due to a change
269 in the definintion of the struct sigaction in
270 glibc 2.0 / libc5. Thanks to Gunther Schlegel
271 <schlegel@riege.de> for reporting the bug and to
272 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
275 2. Modified the label and uuid analysis in order to be
276 self-contained (independant of kernel/libc headers). This
277 should fix the compile with older kernel/libc headers and
278 will preserve the functionality. Thanks to Bernhard Erdmann
279 <bernhard.erdmann@gmx.de> for reporting the bug.
281 3. The 'exclude inode' option, if applied to a directory
282 inode, excludes now correctly all the leaves of this
283 directory. Thanks to John R. Dennison
284 <gerdesas@users.sourceforge.net> for reporting the bug.
286 4. Fixed the '-e' option to disable the possibility
287 to exclude the root inode (which causes the dumps to
288 be unreadable by restore). Prevented array overflow
289 when multiple -e options are used.
291 5. Fixed dump to correctly interpret a filesystem argument
292 which is a mountpoint and it is not an absolute path
293 (as specified in the fstab). Thanks to Bernhard R. Erdmann
294 <be@berdmann.de> for reporting the bug.
296 6. Made dump able to backup files larger than 2 GB. Note that
297 dump still doesn't cope with files larger than 4 GB.
299 7. Restore the real uid and gid before invoking an external
300 RSH program (big hole when dump or restore is suid root!).
302 8. Get the values for the default tape device and the location
303 of fstab file from the system headers. Thanks to
304 Andreas Dilger <adilger@turbolinux.com> for the patch.
306 9. Made dump -w|-W report all recognized filesystems
307 present in either /etc/fstab or /etc/dumpdates, and present
308 the list in the same order as in fstab file. Thanks
309 to Andreas Dilger <adilger@turbolinux.com> for the patch.
311 10. Made dump's -a (automatic end of tape detection) the
312 default. Specifying one of -B, -c, -d or -s options will
313 override the EOM detection. Thanks to Andreas Dilger
314 <adilger@turbolinux.com> for the patch.
316 11. Save the ext2 filesystem label into the dump volume label.
317 Specifying a volume label on the command line (-L option)
318 will override this feature. Thanks to Andreas Dilger
319 <adilger@turbolinux.com> for the patch.
321 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
322 =====================================================================
324 1. Fixed the signal handling in dump (which I broke in 0.4b17)
325 which was causing several strange dump failures (dump
326 hanged or segmentation faults etc).
328 2. Specified the default blocksize in the dump man page.
330 3. Changed two info messages of restore to be written on stdout
331 instead of stderr in order to leave stderr only for errors
332 or warnings. Thanks to Stephen Atwell
333 <satwell@urbana.css.mot.com> for the suggestion.
335 4. Corrected an off by one calculation which prevented
336 dumping a directory having a 1 character name length.
337 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
338 for reporting the bug.
340 5. Reinforce dump in order to allow only level 0 and no
341 -u option when dumping a subdirectory, and document
342 this in the man page. Thanks to Bernhard Erdmann
343 <bernhard.erdmann@gmx.de> for reporting the bug.
345 6. Fixed a small harmless bug in restore which caused
346 in some conditions hard links to be restored several
347 times (therefore generation some warning messages).
348 Thanks to Kenneth Porter <shiva@well.com> for
351 7. Updated the RPM spec file to the latest RedHat version,
352 providing FHS packaging and other cosmetic changes.
353 You will need RPM version >= 3.0.5 in order to build the RPM.
355 8. Updated the configure script to check for libtermcap
356 before checking for libreadline, because we need this
357 library in order to compile the readline support.
359 9. Made dump understand the LABEL= and UUID= notation
360 both in /etc/fstab and on the command line. Note that
361 you will need the /proc filesystem in order to use
362 these notations. Thanks to Erik Troan <ewt@redhat.com>
363 for providing the patch.
365 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
366 ===================================================================
368 1. Fixed a potential buffer overflow in restore. Thanks
369 to Stan Bubrouski <satan@fastdial.net> for reporting
372 2. Fixed a readline-related bug which prevented
373 'cat DUMPFILE | restore -i -f -' from working. Thanks
374 to Charles Karney <karney@users.sourceforge.net>
377 3. Changed a few "panic" into "exit", causing restore to
378 be more stable against some attacks (like the last one
379 reported on Bugtraq, although the last version of restore
380 was not vulnerable - just dumped core). Thanks to
381 Andreas Hasenack <andreas@conectiva.com.br> for reporting
384 4. Removed the suid-root bit on dump and restore in the
385 default build (and generated RPMs). It should be safer
386 now. Users who need the suid-root capabilities in order
387 to make network backups should read first the man page
388 of dump and enable the suid bit manually.
390 5. Added -ltermcap to the compile parameters for restore
391 when using readline support, in order to make the compile
392 process work on some systems (SuSE ?). Thanks to
393 Patrik Schindler <poc@pocnet.net> for reporting the bug.
395 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
396 ====================================================================
398 1. The -F script is called now *only* at the end of a tape,
399 not at the end of the dump. Thanks to Christian Haul
400 <haul@informatik.tu-darmstadt.de> for the bug report.
402 Normally, the device name and the current volume number
403 are passed on the command line of the script. However,
404 if you want the old style script (with no arguments
405 passed on the command line), you can enable it in
406 configure with the --enable-oldstylefscript.
408 2. Use posix signal handling to preserve dump functionality
409 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
412 3. Made the exit code of restore in case of a 'C'ompare
413 command reflect the result of the compare. An exit status
414 of 0 means the dump archive is correct, 1 means tape errors,
415 2 means that some files were modified. Thanks to Kenneth Porter
416 <shiva@well.com> for the suggestion.
418 4. Made (finally) quotes work in interactive restore.
420 5. Small fixes in order to allow dump to compile with a
421 really old e2fsprogs version. Thanks to Ian Zimmerman
422 <itz@speakeasy.org> for the bug report.
424 6. Add GNU readline capabilities to interactive restore.
425 Use configure's --enable-readline flag to enable this feature.
426 Thanks to Patrik Schindler <poc@pocnet.net> for the
429 7. Do the compare on the fly in restore 'C' mode (this will
430 allow not to exhaust the available /tmp space when
431 ccmparing large files). Thanks to Kenneth Porter
432 <shiva@well.com> for the suggestion.
434 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
435 ====================================================================
437 1. Fixed some several duplicate 'DUMP: DUMP:' in the
440 2. Corrected the estimation of blocks to dump. Note that
441 this is still wrong for large directory inodes, due
442 to the size of a BSD directory entry on the tape
443 (estimation is based on the size of the ext2 directory,
444 which is smaller because it doesn't contain the null
445 character at the end).
447 3. Made dump report the total number of MB written to
448 tape at the end of run. Thanks to W. Reilly Cooley
449 <wcooley@nakedape.cc> for the patch.
451 4. Added the -X option to restore giving the possibility
452 to read the names of the files to be extracted/listed
453 from a text file (in addition of the command line).
454 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
457 5. Added the device name and the current volume number
458 as arguments to the end of tape script (dump -F option).
460 6. Made the multi-volume dump work again (after having
461 broken it in 0.4b15).
463 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
464 ===================================================================
466 1. Added a prompt command in interactive restore mode. Thanks
467 to Andreas Dilger <adilger@home.com> for the patch.
469 2. Fixed a buffer overflow problem in dump (caused by
470 not checking the size of the filesystem parameter).
471 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
472 reporting this on Bugtraq (and to several dump users
473 who forwarded me his mail).
475 3. Added the '-F script' option to dump in order to
476 launch a script at the end of each tape (to be used
477 with a tape changer, or to notify the sysadmin by
480 4. Fixed a bug in restore compare code caused by the changes
483 5. Fixed the treatment of options using the old BSD syntax
484 in both dump and restore.
486 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
487 =======================================================================
489 1. Fixed a bug in dump which may have caused invalid deleted
490 directories to be dumped out if they were deleted after the
491 mapping pass. This could occure on active filesystem and lead
492 to heap corruption (causing dump malfunction in many possible ways).
493 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
494 bug and submitting the patch.
496 2. Fixed the handling of the filesystem parameter in dump. A
497 '/mnt/disk' parameter caused the disk contents to be dumped,
498 but a '/mnt/disk/' parameter caused the mountpoint directory
499 to be dumped (generally an empty directory).
501 3. Improved the output of dump in order to tell which directory
502 it is currently dumping (when dumping a subtree).
504 4. Added the '-e' exclude inode option to dump. Thanks to
505 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
507 5. Added a REPORTING-BUGS file in order to provide a guide
508 on how to correctly report a bug in dump/restore.
510 6. Fixed a restore problem when restoring a hard link to an inode
511 having the immutable or append-only attribute set. Thanks to
512 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
514 7. Fixed a compatibility problem between dumps made on little
515 endian machines (the format was correct) and big endian
516 machines (the format was incorrect). This fix break the
517 compatibility with the older dumps made on big endian
518 machines (sparc, m86k, ppc etc). For the first time in
519 linux dump's history (I believe), the dumps made by *BSD,
520 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
521 so interchangeable. Thanks to Rob Cermak
522 <cermak@ahab.rutgers.edu> for submitting the bug and
523 helping me test the fix.
525 8. Fixed the way dump reports the remaining percent/time, if
526 the number of blocks actually dumped exceeds the estimated
527 number of blocks. Thanks to Jean-Paul van der Jagt
528 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
530 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
531 ======================================================================
533 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
534 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
536 2. Removed the requirement to build the RPM as root from the
537 spec file. Thanks to Christian Weisgerber
538 <naddy@mips.rhein-neckar.de> for submitting this.
540 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
541 causing dump to block when dumping really huge filesystems.
542 Many thanks to Patrik Schindler <poc@pocnet.net> for
543 helping me find this bug.
545 4. Fixed the treatment for an interrupt signal when dump access
546 the remote tape through RSH. Thanks to Christian Weisgerber
547 <naddy@mips.rhein-neckar.de> for providing the patch.
549 5. Fixed a bug which was causing dump/restore to display
550 garbage characters instead of the remote host name.
552 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
553 =====================================================================
555 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
556 <kukuk@suse.de> for submitting the patch.
558 2. Fix for the exit code when using the size estimate option of
559 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
560 submitting the patch.
562 3. Handle EINTR in atomical reads/writes in dump, which was causing
563 dump to fail on some systems. Thanks to Eric Jergensen
564 <eric@dvns.com> for reporting the bug and submitting the patch.
566 4. Handle more than 16 characters for the device names in dumpdates.
567 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
568 tracking down the problem and proposing the solution.
570 5. Fixed a bug in dump which prevented the creation of the
571 dumpdates file when doing a 0-level dump without already
572 having a dumpdates file. Thanks to Patrik Schindler
573 <poc@pocnet.net> for reporting the bug.
575 6. Changed the way dump 'S' flag reports the size estimate
576 from number of blocks into bytes (making it compatible
577 with the Solaris version, and simplifying things for
578 amanda users). Thanks to Jason L Tibbitts III
579 <tibbs@math.uh.edu> for reporting the bug.
581 7. Fixed a compatibility problem in linux/alpha dump tape format.
582 Now the linux/alpha dump are (again) compatible with the
583 other dump formats. But this breaks compatibility with
584 older dumps made on alpha. Thanks to Mike Tibor
585 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
587 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
588 ======================================================================
590 1. Added a '--enable-kerberos' to configure.
592 2. Added a 'S' option to dump which determines the amount of space
593 that is needed to perform the dump without actually doing it, similar
594 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
595 <cermak@ahab.rutgers.edu>.
597 3. Added a 'M' multi-volume option to dump and restore which enables
598 dumping to multiple files (useful when dumping to an ext2
599 partition to make several dump files in order to bypass the 2GB
600 file size limitation). The 'f' argument is treated as a prefix and
601 the output files will be named <prefix>001, <prefix>002 etc. With
602 the 'M' flag, restore automatically selects the right file without
603 asking to enter a new tape each time.
605 4. Fixed a memory leak which was causing dump to grow very big
606 (270MB when dumping a 10GB filesystem...). Thanks to Jason
607 Fearon <jasonf@netrider.org.au> for reporting the bug.
609 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
610 ======================================================================
612 1. Make configure test if the system glob routines support
613 extended syntax (ALTDIRFUNC). If not, use the internal glob
614 routines instead of system ones. Thanks to Bernhard Sadlowski
615 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
616 and helping me resolve this and other minor libc5 compiling
619 2. Fix a problem when dumping a ext2fs with the 'filetype'
620 feature enabled. Thanks to Patrick J. LoPresti
621 <patl@cag.lcs.mit.edu> for reporting the bug and to
622 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
624 3. Made the nodump flag work on directories. A directory which
625 has the nodump flag gets never dumped, regardless of its
628 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
629 which allows dump on an active ext3 filesystem. However, this
630 is a "quick and dirty" patch which enables backup of an ext3
631 filesystem through the ext2 compatibility (by ignoring the
632 NEEDS_RECOVERY bit). The journal file is not recognized and
633 it is dumped (it should not).
635 5. Test the superblock compatibility flags when dumping, in order
636 to be sure that we know how to deal with specific features.
638 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
639 ====================================================================
641 1. Use lchown instead of chown, fixing a possible security problem
642 when restoring symlinks (a malicious user could use this
643 to deliberately corrupt the ownership of important system files).
644 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
645 this and providing the patch.
647 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
648 ====================================================================
650 1. Put dump sources under CVS, added Id tags in all files so
651 one can use 'ident' on binary files.
653 2. Added the dump/restore version in the usage text so one can
654 easily verify the version he is using.
656 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
657 a va_start/va_end problem on linux-ppc (always call va_start
658 va_end in pairs each time we use a vararg function).
660 4. Added again the DT_* constants because old libc does not
661 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
662 for submitting the bug report.
664 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
665 this should enable dumping big (huge) filesystems.
667 6. Added the RSH environment variable in order to be able to
668 use a rsh replacement like ssh when doing remote backups (and
669 bypass the security limitations of rcmd). Now you can do remote
670 backups without being root (or making dump setuid root).
672 7. Modified again the way dumpdates works. For incremental dumps,
673 we need to read dumpdates even if we are not using 'u' option.
674 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
677 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
678 ===================================================================
680 1. Removed the 'k' flag from the restore 'about' text if kerberos
683 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
684 to them (fsetflags takes a char*, setflags an open fd!).
686 3. (f)setflags is called only if the flags aren't empty. If the
687 file is a special file, a warning is printed, because changing
688 flags implies opening the device. Normally, a special file
689 should not have any flag... (Debian bug #29775, patch provided
690 by Abhijit Dasgupta <abhijit@ans.net>).
692 4. Made possible to dump a file system not mentioned in /etc/fstab.
693 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
695 5. Changed the default behaviour to not create dumpdates
696 unless 'u' option is specified. Removed the old "debian-patch"
697 which provided the same thing. (Debian bug #38136, #33818).
699 6. Removed all those dump*announce, since they were getting old...
701 7. Added warning messages if dumpdates does not exist and
702 when an inferior level dump does not exist (except for a level 0
705 8. Debugged the glob calls in interactive mode: restore used a
706 dirent struct which was different from the /usr/include/dirent.h
707 one (this used to work, is it a glibc2 change?), so none of the
708 compat glob (which used /usr/include/dirent.h) or the system glob
709 worked. Restore use now the system dirent (and the system
710 DT_* constants), which are compatible with BSD ones.
712 9. Added a configure flag (--with-dumpdatespath) to specify
713 the location of dumpdates file. By default, it is
716 10. Added the "AUTHOR" and "AVAILABILITY" sections and
717 included the current date/version in man pages.
719 11. Corrected the estimation of remaining time when
720 the operator doesn't change the tapes quickly enough. This
721 was an old bug, I thought I corrected it, and discovered
722 that in fact it was corrected in two different places, so
723 the results canceled each other...
725 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
726 ===================================================================
728 1. Integrated multiple patches from RedHat, Debian and SuSE:
730 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
732 - several fixes in the man pages.
733 - update the default tape device to /dev/st0.
734 - many updates for Linux Alpha (byte ordering, size_t etc).
736 - use environment variable for TMPDIR (instead of /tmp).
737 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
738 - workaround egcs bug (RedHat bugs #4281 and #2989).
739 - wire $(OPT) throughout Makefile's.
741 2. Upgrade the dump revision to 1, making possible to dump filesystems
742 made with e2fsprogs-1.15 or newer. Nothing seems to break...
744 3. Fix some compile warnings, prototype all functions.
746 4. Use glibc err/glob instead of internal compatibility
747 routines (only if available).
749 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
750 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
752 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
753 ======================================================================
755 1. Integrated the changes from FreeBSD-3.1-RELEASE
756 (mostly bug fixes, buffer overruns, dump has now an "automatic
757 tape length calculation" flag, dump/restore can use kerberos now
758 (this is NOT tested), use environment variables for TAPE and
761 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
763 3. Corrected a bug in restore when using 'C' option with multi-volumes
764 tapes (files splited accros two tapes give "size changed" errors
767 4. Corrected the long standing bug when dumping multiple tapes.
768 This works for me, needs further testing.
770 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
771 ====================================================================
773 1. Dump now runs correctly on kernels 2.1.x
774 Fix made by Gerald Peters <gapeters@worldnet.att.net>
776 Changes between versions 0.4b2 and 0.4b3
777 ========================================
779 1. Use realpath() if available
783 Changes between versions 0.4b1 and 0.4b2
784 ========================================
786 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
789 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
791 3. Fixed configure.in to correctly handle --enable-rmt
793 Changes between versions 0.3 and 0.4b1
794 ======================================
796 1. Integrated the changes from 4.4BSD-Lite2
798 2. Integrated the patches from Debian and Red Hat
800 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
802 4. Changed dump to use the Ext2fs library to get block addresses. This
803 should solve the endianness problem on SparcLinux.
805 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
806 distribution's one) to use autoconf
808 6. Fixed a few minor bugs
810 Changes between versions 0.2e and 0.2f
811 ======================================
813 1. Added the creation of named pipes (fifos) in restore.
815 2. Added the -N flag in the restore manual page.
817 3. Added the file kernel.patch which contains the llseek() optimization
818 patch for 1.2.x kernels.
820 4. Fixed a bug in the restoration of symbolic links: owner and group were
823 5. Integrated some changes from FreeBSD 2.2.
825 6. Added a call to ftruncate() after restoring each file to restore
826 correctly files ending by a hole.
828 Changes between versions 0.2d and 0.2e
829 ======================================
831 1. Fixed a bug in the "set owner/mode" process. Every file was restored
832 with owner = root (0) and group = root/wheel/whatever (0).
834 Changes between versions 0.2c and 0.2d
835 ======================================
837 1. Dump is now able to backup 2GB+ filesystems.
839 2. Dump and restore can now be linked as static binaries.
841 Changes between versions 0.2b and 0.2c
842 ======================================
844 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
846 Changes between versions 0.2a and 0.2b
847 ======================================
849 1. Really fixed the bug that I should have corrected in 0.2a.
851 2. Enabled optimization again.
853 Changes between versions 0.2 and 0.2a
854 =====================================
856 1. Disabled the optimization during compilation.
858 Changes between versions 0.1 and 0.2
859 ====================================
861 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
862 the fs_type field (actually, I modified the file fstab.c to make it
863 use the mntent functions).
865 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
866 made by dump should be readable by the BSD restore and backups made
867 by the BSD dump should be readable by restore. Unfortunately, this
868 means that the dump format has changed between version 0.1 and version
871 3. Dump is now able to backup a subtree, it is no longer limited to whole
872 filesystems like the BSD version.
874 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
877 Changes between versions 0.0 and 0.1
878 ====================================
880 1. Now create links rdump and rrestore during the `make install' step.
882 2. Linux port specific bugs added to the manual pages
884 3. Incorrect estimation of the number of tapes blocks fixed when doing
887 4. Better ls-like format in restore in interactive mode.