1 $Id: CHANGES,v 1.116 2001/05/26 11:11:16 stelian Exp $
3 Changes between versions 0.4b22 and 0.4b23 (released ????????????)
4 ==================================================================
6 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
7 Marcus Meissner (Caldera International Security Dept.).
9 2. Implement the Sun rmt extended protocol. Patch submitted
10 by Ian Gordon <iangordon@users.sourceforge.net>.
12 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
13 ==================================================================
15 1. Made dump report the number of blocks dumped per volume.
16 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
18 2. Fix a bug in dump where waiting too much at the 'change volumes'
19 question caused the volume to be rewritten. Thanks to
20 Graham Reed <greed@users.sourceforge.net> for reporting the
21 bug and providing a patch.
23 3. Added a compression option to dump, contributed by Jerry
24 Peters <gapeters@worldnet.att.net>.
26 WARNING: the compression works only on files/pipes and
27 drives supporting variable block size.
29 WARNING2: tapes created using the compression feature are
30 incompatible with the standard dump tape format, and a
31 version of restore >= 0.4b22 will be needed for extraction.
33 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
34 architectures. Thanks to Paul Slootman <paul@debian.org> for
35 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
38 5. Many cleanups (CPP defines, const char warnings, check of
39 ext2fs COMPAT flags, time_t cleanups, added libext2 version
40 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
42 6. Made --prefix option work in configure. All the install path
43 are now based on the configure parameters.
45 7. Added the Quick File Access mode in dump/restore, contributed
46 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
47 a file tape position for each inode, and this file is used by
48 restore (if called with parameter Q and the filename)
49 to directly position the tape at the file restore is currently
50 working on. This saves hours when restoring single files from
51 large backups, saves the tapes and the drive's head. Use
52 --enable-qfa option of configure to compile in the QFA support.
54 8. Added the possibility to dump several files and directories
55 in a single invocation of dump. Thanks to Uwe Gohlke
56 <uwe@ugsoft.de> for implementing this option.
58 9. Fixed the dumping and restoring of files with holes
59 (files ending with holes didn't get dumped or restored
62 10. Fixed a socket descriptor leak in dump, which leaved opened
63 3 file descriptors per dump process (and there is one dump
66 11. Fixed dump large file system support, by explicit use of
67 open64/lseek64/etc functions (explicit use needed because
68 e2fsprogs libraries don't behave well when compiled with
71 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
72 ======================================================================
74 1. Fixed some bugs in the dump -w|-W logic introduced by
75 the previous version. Thanks to Andreas Dilger
76 <adilger@turbolinux.com> for his help on this one.
78 2. Fixed again a compilation problem when using old e2fs
79 headers (filesystem label related). Thanks to many users
80 who reported this stupid error.
82 3. Fixed a build problem on old lib5 based systems dealing with
83 _PATH_MNTTAB being undefined. Thanks to John Adams
84 <johna@onevista.com> for reporting the bug.
86 4. Improved the error detection in restore when creating
87 the temporary files in TMPDIR. Now restore will corectly
88 report a 'no space left on device' error instead of
89 strange errors which could imply an archive corruption.
90 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
91 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
93 5. Added the throughput information in dump's progression
94 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
97 6. Use libext2fs's inode scanning routines, which are particularly
98 robust when dealing with errors (especially when having some
99 bad blocks on the filesystem). Many thanks to Colin
100 <colin@horizon.com> for reporting the bug, investigating
101 the issues, finding a workaround, writing the patch and
102 fully testing it... (of course, if this patch will break
103 anything, he is to blame for :-)).
105 7. Made dump and restore LFS aware. Dump can dump a filesystem
106 containing large files, generate a large file on output and
107 restore can restore them. This feature is not enabled by
108 default, use --enable-largefile option of configure to enable
109 it (you need to have a LFS aware glibc though). Thanks to
110 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
111 and to Theodore T'so <tytso@valinux.com> for his always
114 8. Made dump ask upon a tape write error if it should rewrite
115 the current volume (assume this is a bad tape) or if it should
116 assume an end-of-tape condition (useful for tape drives which
117 behaves badly at the end of the tape). Thanks to Andreas
118 Dilger <adilger@turbolinux.com> for the suggestion.
120 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
121 =======================================================================
123 1. Fixed a small compilation problem due to a change
124 in the definintion of the struct sigaction in
125 glibc 2.0 / libc5. Thanks to Gunther Schlegel
126 <schlegel@riege.de> for reporting the bug and to
127 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
130 2. Modified the label and uuid analysis in order to be
131 self-contained (independant of kernel/libc headers). This
132 should fix the compile with older kernel/libc headers and
133 will preserve the functionality. Thanks to Bernhard Erdmann
134 <bernhard.erdmann@gmx.de> for reporting the bug.
136 3. The 'exclude inode' option, if applied to a directory
137 inode, excludes now correctly all the leaves of this
138 directory. Thanks to John R. Dennison
139 <gerdesas@users.sourceforge.net> for reporting the bug.
141 4. Fixed the '-e' option to disable the possibility
142 to exclude the root inode (which causes the dumps to
143 be unreadable by restore). Prevented array overflow
144 when multiple -e options are used.
146 5. Fixed dump to correctly interpret a filesystem argument
147 which is a mountpoint and it is not an absolute path
148 (as specified in the fstab). Thanks to Bernhard R. Erdmann
149 <be@berdmann.de> for reporting the bug.
151 6. Made dump able to backup files larger than 2 GB. Note that
152 dump still doesn't cope with files larger than 4 GB.
154 7. Restore the real uid and gid before invoking an external
155 RSH program (big hole when dump or restore is suid root!).
157 8. Get the values for the default tape device and the location
158 of fstab file from the system headers. Thanks to
159 Andreas Dilger <adilger@turbolinux.com> for the patch.
161 9. Made dump -w|-W report all recognized filesystems
162 present in either /etc/fstab or /etc/dumpdates, and present
163 the list in the same order as in fstab file. Thanks
164 to Andreas Dilger <adilger@turbolinux.com> for the patch.
166 10. Made dump's -a (automatic end of tape detection) the
167 default. Specifying one of -B, -c, -d or -s options will
168 override the EOM detection. Thanks to Andreas Dilger
169 <adilger@turbolinux.com> for the patch.
171 11. Save the ext2 filesystem label into the dump volume label.
172 Specifying a volume label on the command line (-L option)
173 will override this feature. Thanks to Andreas Dilger
174 <adilger@turbolinux.com> for the patch.
176 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
177 =====================================================================
179 1. Fixed the signal handling in dump (which I broke in 0.4b17)
180 which was causing several strange dump failures (dump
181 hanged or segmentation faults etc).
183 2. Specified the default blocksize in the dump man page.
185 3. Changed two info messages of restore to be written on stdout
186 instead of stderr in order to leave stderr only for errors
187 or warnings. Thanks to Stephen Atwell
188 <satwell@urbana.css.mot.com> for the suggestion.
190 4. Corrected an off by one calculation which prevented
191 dumping a directory having a 1 character name length.
192 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
193 for reporting the bug.
195 5. Reinforce dump in order to allow only level 0 and no
196 -u option when dumping a subdirectory, and document
197 this in the man page. Thanks to Bernhard Erdmann
198 <bernhard.erdmann@gmx.de> for reporting the bug.
200 6. Fixed a small harmless bug in restore which caused
201 in some conditions hard links to be restored several
202 times (therefore generation some warning messages).
203 Thanks to Kenneth Porter <shiva@well.com> for
206 7. Updated the RPM spec file to the latest RedHat version,
207 providing FHS packaging and other cosmetic changes.
208 You will need RPM version >= 3.0.5 in order to build the RPM.
210 8. Updated the configure script to check for libtermcap
211 before checking for libreadline, because we need this
212 library in order to compile the readline support.
214 9. Made dump understand the LABEL= and UUID= notation
215 both in /etc/fstab and on the command line. Note that
216 you will need the /proc filesystem in order to use
217 these notations. Thanks to Erik Troan <ewt@redhat.com>
218 for providing the patch.
220 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
221 ===================================================================
223 1. Fixed a potential buffer overflow in restore. Thanks
224 to Stan Bubrouski <satan@fastdial.net> for reporting
227 2. Fixed a readline-related bug which prevented
228 'cat DUMPFILE | restore -i -f -' from working. Thanks
229 to Charles Karney <karney@users.sourceforge.net>
232 3. Changed a few "panic" into "exit", causing restore to
233 be more stable against some attacks (like the last one
234 reported on Bugtraq, although the last version of restore
235 was not vulnerable - just dumped core). Thanks to
236 Andreas Hasenack <andreas@conectiva.com.br> for reporting
239 4. Removed the suid-root bit on dump and restore in the
240 default build (and generated RPMs). It should be safer
241 now. Users who need the suid-root capabilities in order
242 to make network backups should read first the man page
243 of dump and enable the suid bit manually.
245 5. Added -ltermcap to the compile parameters for restore
246 when using readline support, in order to make the compile
247 process work on some systems (SuSE ?). Thanks to
248 Patrik Schindler <poc@pocnet.net> for reporting the bug.
250 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
251 ====================================================================
253 1. The -F script is called now *only* at the end of a tape,
254 not at the end of the dump. Thanks to Christian Haul
255 <haul@informatik.tu-darmstadt.de> for the bug report.
257 Normally, the device name and the current volume number
258 are passed on the command line of the script. However,
259 if you want the old style script (with no arguments
260 passed on the command line), you can enable it in
261 configure with the --enable-oldstylefscript.
263 2. Use posix signal handling to preserve dump functionality
264 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
267 3. Made the exit code of restore in case of a 'C'ompare
268 command reflect the result of the compare. An exit status
269 of 0 means the dump archive is correct, 1 means tape errors,
270 2 means that some files were modified. Thanks to Kenneth Porter
271 <shiva@well.com> for the suggestion.
273 4. Made (finally) quotes work in interactive restore.
275 5. Small fixes in order to allow dump to compile with a
276 really old e2fsprogs version. Thanks to Ian Zimmerman
277 <itz@speakeasy.org> for the bug report.
279 6. Add GNU readline capabilities to interactive restore.
280 Use configure's --enable-readline flag to enable this feature.
281 Thanks to Patrik Schindler <poc@pocnet.net> for the
284 7. Do the compare on the fly in restore 'C' mode (this will
285 allow not to exhaust the available /tmp space when
286 ccmparing large files). Thanks to Kenneth Porter
287 <shiva@well.com> for the suggestion.
289 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
290 ====================================================================
292 1. Fixed some several duplicate 'DUMP: DUMP:' in the
295 2. Corrected the estimation of blocks to dump. Note that
296 this is still wrong for large directory inodes, due
297 to the size of a BSD directory entry on the tape
298 (estimation is based on the size of the ext2 directory,
299 which is smaller because it doesn't contain the null
300 character at the end).
302 3. Made dump report the total number of MB written to
303 tape at the end of run. Thanks to W. Reilly Cooley
304 <wcooley@nakedape.cc> for the patch.
306 4. Added the -X option to restore giving the possibility
307 to read the names of the files to be extracted/listed
308 from a text file (in addition of the command line).
309 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
312 5. Added the device name and the current volume number
313 as arguments to the end of tape script (dump -F option).
315 6. Made the multi-volume dump work again (after having
316 broken it in 0.4b15).
318 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
319 ===================================================================
321 1. Added a prompt command in interactive restore mode. Thanks
322 to Andreas Dilger <adilger@home.com> for the patch.
324 2. Fixed a buffer overflow problem in dump (caused by
325 not checking the size of the filesystem parameter).
326 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
327 reporting this on Bugtraq (and to several dump users
328 who forwarded me his mail).
330 3. Added the '-F script' option to dump in order to
331 launch a script at the end of each tape (to be used
332 with a tape changer, or to notify the sysadmin by
335 4. Fixed a bug in restore compare code caused by the changes
338 5. Fixed the treatment of options using the old BSD syntax
339 in both dump and restore.
341 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
342 =======================================================================
344 1. Fixed a bug in dump which may have caused invalid deleted
345 directories to be dumped out if they were deleted after the
346 mapping pass. This could occure on active filesystem and lead
347 to heap corruption (causing dump malfunction in many possible ways).
348 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
349 bug and submitting the patch.
351 2. Fixed the handling of the filesystem parameter in dump. A
352 '/mnt/disk' parameter caused the disk contents to be dumped,
353 but a '/mnt/disk/' parameter caused the mountpoint directory
354 to be dumped (generally an empty directory).
356 3. Improved the output of dump in order to tell which directory
357 it is currently dumping (when dumping a subtree).
359 4. Added the '-e' exclude inode option to dump. Thanks to
360 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
362 5. Added a REPORTING-BUGS file in order to provide a guide
363 on how to correctly report a bug in dump/restore.
365 6. Fixed a restore problem when restoring a hard link to an inode
366 having the immutable or append-only attribute set. Thanks to
367 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
369 7. Fixed a compatibility problem between dumps made on little
370 endian machines (the format was correct) and big endian
371 machines (the format was incorrect). This fix break the
372 compatibility with the older dumps made on big endian
373 machines (sparc, m86k, ppc etc). For the first time in
374 linux dump's history (I believe), the dumps made by *BSD,
375 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
376 so interchangeable. Thanks to Rob Cermak
377 <cermak@ahab.rutgers.edu> for submitting the bug and
378 helping me test the fix.
380 8. Fixed the way dump reports the remaining percent/time, if
381 the number of blocks actually dumped exceeds the estimated
382 number of blocks. Thanks to Jean-Paul van der Jagt
383 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
385 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
386 ======================================================================
388 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
389 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
391 2. Removed the requirement to build the RPM as root from the
392 spec file. Thanks to Christian Weisgerber
393 <naddy@mips.rhein-neckar.de> for submitting this.
395 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
396 causing dump to block when dumping really huge filesystems.
397 Many thanks to Patrik Schindler <poc@pocnet.net> for
398 helping me find this bug.
400 4. Fixed the treatment for an interrupt signal when dump access
401 the remote tape through RSH. Thanks to Christian Weisgerber
402 <naddy@mips.rhein-neckar.de> for providing the patch.
404 5. Fixed a bug which was causing dump/restore to display
405 garbage characters instead of the remote host name.
407 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
408 =====================================================================
410 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
411 <kukuk@suse.de> for submitting the patch.
413 2. Fix for the exit code when using the size estimate option of
414 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
415 submitting the patch.
417 3. Handle EINTR in atomical reads/writes in dump, which was causing
418 dump to fail on some systems. Thanks to Eric Jergensen
419 <eric@dvns.com> for reporting the bug and submitting the patch.
421 4. Handle more than 16 characters for the device names in dumpdates.
422 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
423 tracking down the problem and proposing the solution.
425 5. Fixed a bug in dump which prevented the creation of the
426 dumpdates file when doing a 0-level dump without already
427 having a dumpdates file. Thanks to Patrik Schindler
428 <poc@pocnet.net> for reporting the bug.
430 6. Changed the way dump 'S' flag reports the size estimate
431 from number of blocks into bytes (making it compatible
432 with the Solaris version, and simplifying things for
433 amanda users). Thanks to Jason L Tibbitts III
434 <tibbs@math.uh.edu> for reporting the bug.
436 7. Fixed a compatibility problem in linux/alpha dump tape format.
437 Now the linux/alpha dump are (again) compatible with the
438 other dump formats. But this breaks compatibility with
439 older dumps made on alpha. Thanks to Mike Tibor
440 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
442 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
443 ======================================================================
445 1. Added a '--enable-kerberos' to configure.
447 2. Added a 'S' option to dump which determines the amount of space
448 that is needed to perform the dump without actually doing it, similar
449 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
450 <cermak@ahab.rutgers.edu>.
452 3. Added a 'M' multi-volume option to dump and restore which enables
453 dumping to multiple files (useful when dumping to an ext2
454 partition to make several dump files in order to bypass the 2GB
455 file size limitation). The 'f' argument is treated as a prefix and
456 the output files will be named <prefix>001, <prefix>002 etc. With
457 the 'M' flag, restore automatically selects the right file without
458 asking to enter a new tape each time.
460 4. Fixed a memory leak which was causing dump to grow very big
461 (270MB when dumping a 10GB filesystem...). Thanks to Jason
462 Fearon <jasonf@netrider.org.au> for reporting the bug.
464 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
465 ======================================================================
467 1. Make configure test if the system glob routines support
468 extended syntax (ALTDIRFUNC). If not, use the internal glob
469 routines instead of system ones. Thanks to Bernhard Sadlowski
470 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
471 and helping me resolve this and other minor libc5 compiling
474 2. Fix a problem when dumping a ext2fs with the 'filetype'
475 feature enabled. Thanks to Patrick J. LoPresti
476 <patl@cag.lcs.mit.edu> for reporting the bug and to
477 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
479 3. Made the nodump flag work on directories. A directory which
480 has the nodump flag gets never dumped, regardless of its
483 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
484 which allows dump on an active ext3 filesystem. However, this
485 is a "quick and dirty" patch which enables backup of an ext3
486 filesystem through the ext2 compatibility (by ignoring the
487 NEEDS_RECOVERY bit). The journal file is not recognized and
488 it is dumped (it should not).
490 5. Test the superblock compatibility flags when dumping, in order
491 to be sure that we know how to deal with specific features.
493 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
494 ====================================================================
496 1. Use lchown instead of chown, fixing a possible security problem
497 when restoring symlinks (a malicious user could use this
498 to deliberately corrupt the ownership of important system files).
499 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
500 this and providing the patch.
502 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
503 ====================================================================
505 1. Put dump sources under CVS, added Id tags in all files so
506 one can use 'ident' on binary files.
508 2. Added the dump/restore version in the usage text so one can
509 easily verify the version he is using.
511 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
512 a va_start/va_end problem on linux-ppc (always call va_start
513 va_end in pairs each time we use a vararg function).
515 4. Added again the DT_* constants because old libc does not
516 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
517 for submitting the bug report.
519 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
520 this should enable dumping big (huge) filesystems.
522 6. Added the RSH environment variable in order to be able to
523 use a rsh replacement like ssh when doing remote backups (and
524 bypass the security limitations of rcmd). Now you can do remote
525 backups without being root (or making dump setuid root).
527 7. Modified again the way dumpdates works. For incremental dumps,
528 we need to read dumpdates even if we are not using 'u' option.
529 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
532 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
533 ===================================================================
535 1. Removed the 'k' flag from the restore 'about' text if kerberos
538 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
539 to them (fsetflags takes a char*, setflags an open fd!).
541 3. (f)setflags is called only if the flags aren't empty. If the
542 file is a special file, a warning is printed, because changing
543 flags implies opening the device. Normally, a special file
544 should not have any flag... (Debian bug #29775, patch provided
545 by Abhijit Dasgupta <abhijit@ans.net>).
547 4. Made possible to dump a file system not mentioned in /etc/fstab.
548 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
550 5. Changed the default behaviour to not create dumpdates
551 unless 'u' option is specified. Removed the old "debian-patch"
552 which provided the same thing. (Debian bug #38136, #33818).
554 6. Removed all those dump*announce, since they were getting old...
556 7. Added warning messages if dumpdates does not exist and
557 when an inferior level dump does not exist (except for a level 0
560 8. Debugged the glob calls in interactive mode: restore used a
561 dirent struct which was different from the /usr/include/dirent.h
562 one (this used to work, is it a glibc2 change?), so none of the
563 compat glob (which used /usr/include/dirent.h) or the system glob
564 worked. Restore use now the system dirent (and the system
565 DT_* constants), which are compatible with BSD ones.
567 9. Added a configure flag (--with-dumpdatespath) to specify
568 the location of dumpdates file. By default, it is
571 10. Added the "AUTHOR" and "AVAILABILITY" sections and
572 included the current date/version in man pages.
574 11. Corrected the estimation of remaining time when
575 the operator doesn't change the tapes quickly enough. This
576 was an old bug, I thought I corrected it, and discovered
577 that in fact it was corrected in two different places, so
578 the results canceled each other...
580 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
581 ===================================================================
583 1. Integrated multiple patches from RedHat, Debian and SuSE:
585 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
587 - several fixes in the man pages.
588 - update the default tape device to /dev/st0.
589 - many updates for Linux Alpha (byte ordering, size_t etc).
591 - use environment variable for TMPDIR (instead of /tmp).
592 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
593 - workaround egcs bug (RedHat bugs #4281 and #2989).
594 - wire $(OPT) throughout Makefile's.
596 2. Upgrade the dump revision to 1, making possible to dump filesystems
597 made with e2fsprogs-1.15 or newer. Nothing seems to break...
599 3. Fix some compile warnings, prototype all functions.
601 4. Use glibc err/glob instead of internal compatibility
602 routines (only if available).
604 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
605 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
607 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
608 ======================================================================
610 1. Integrated the changes from FreeBSD-3.1-RELEASE
611 (mostly bug fixes, buffer overruns, dump has now an "automatic
612 tape length calculation" flag, dump/restore can use kerberos now
613 (this is NOT tested), use environment variables for TAPE and
616 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
618 3. Corrected a bug in restore when using 'C' option with multi-volumes
619 tapes (files splited accros two tapes give "size changed" errors
622 4. Corrected the long standing bug when dumping multiple tapes.
623 This works for me, needs further testing.
625 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
626 ====================================================================
628 1. Dump now runs correctly on kernels 2.1.x
629 Fix made by Gerald Peters <gapeters@worldnet.att.net>
631 Changes between versions 0.4b2 and 0.4b3
632 ========================================
634 1. Use realpath() if available
638 Changes between versions 0.4b1 and 0.4b2
639 ========================================
641 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
644 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
646 3. Fixed configure.in to correctly handle --enable-rmt
648 Changes between versions 0.3 and 0.4b1
649 ======================================
651 1. Integrated the changes from 4.4BSD-Lite2
653 2. Integrated the patches from Debian and Red Hat
655 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
657 4. Changed dump to use the Ext2fs library to get block addresses. This
658 should solve the endianness problem on SparcLinux.
660 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
661 distribution's one) to use autoconf
663 6. Fixed a few minor bugs
665 Changes between versions 0.2e and 0.2f
666 ======================================
668 1. Added the creation of named pipes (fifos) in restore.
670 2. Added the -N flag in the restore manual page.
672 3. Added the file kernel.patch which contains the llseek() optimization
673 patch for 1.2.x kernels.
675 4. Fixed a bug in the restoration of symbolic links: owner and group were
678 5. Integrated some changes from FreeBSD 2.2.
680 6. Added a call to ftruncate() after restoring each file to restore
681 correctly files ending by a hole.
683 Changes between versions 0.2d and 0.2e
684 ======================================
686 1. Fixed a bug in the "set owner/mode" process. Every file was restored
687 with owner = root (0) and group = root/wheel/whatever (0).
689 Changes between versions 0.2c and 0.2d
690 ======================================
692 1. Dump is now able to backup 2GB+ filesystems.
694 2. Dump and restore can now be linked as static binaries.
696 Changes between versions 0.2b and 0.2c
697 ======================================
699 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
701 Changes between versions 0.2a and 0.2b
702 ======================================
704 1. Really fixed the bug that I should have corrected in 0.2a.
706 2. Enabled optimization again.
708 Changes between versions 0.2 and 0.2a
709 =====================================
711 1. Disabled the optimization during compilation.
713 Changes between versions 0.1 and 0.2
714 ====================================
716 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
717 the fs_type field (actually, I modified the file fstab.c to make it
718 use the mntent functions).
720 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
721 made by dump should be readable by the BSD restore and backups made
722 by the BSD dump should be readable by restore. Unfortunately, this
723 means that the dump format has changed between version 0.1 and version
726 3. Dump is now able to backup a subtree, it is no longer limited to whole
727 filesystems like the BSD version.
729 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
732 Changes between versions 0.0 and 0.1
733 ====================================
735 1. Now create links rdump and rrestore during the `make install' step.
737 2. Linux port specific bugs added to the manual pages
739 3. Incorrect estimation of the number of tapes blocks fixed when doing
742 4. Better ls-like format in restore in interactive mode.