]> git.wh0rd.org - dump.git/blob - CHANGES
Dump -w|-W fixes
[dump.git] / CHANGES
1 $Id: CHANGES,v 1.91 2000/11/29 10:13:41 stelian Exp $
2
3 Changes between versions 0.4b20 and 0.4b21 (released ?????????????????)
4 =======================================================================
5
6 1. Fixed some bugs in the dump -w|-W logic introduced by
7 the previous version. Thanks to Andreas Dilger
8 <adilger@turbolinux.com> for his help on this one.
9
10 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
11 =======================================================================
12
13 1. Fixed a small compilation problem due to a change
14 in the definintion of the struct sigaction in
15 glibc 2.0 / libc5. Thanks to Gunther Schlegel
16 <schlegel@riege.de> for reporting the bug and to
17 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
18 a fix.
19
20 2. Modified the label and uuid analysis in order to be
21 self-contained (independant of kernel/libc headers). This
22 should fix the compile with older kernel/libc headers and
23 will preserve the functionality. Thanks to Bernhard Erdmann
24 <bernhard.erdmann@gmx.de> for reporting the bug.
25
26 3. The 'exclude inode' option, if applied to a directory
27 inode, excludes now correctly all the leaves of this
28 directory. Thanks to John R. Dennison
29 <gerdesas@users.sourceforge.net> for reporting the bug.
30
31 4. Fixed the '-e' option to disable the possibility
32 to exclude the root inode (which causes the dumps to
33 be unreadable by restore). Prevented array overflow
34 when multiple -e options are used.
35
36 5. Fixed dump to correctly interpret a filesystem argument
37 which is a mountpoint and it is not an absolute path
38 (as specified in the fstab). Thanks to Bernhard R. Erdmann
39 <be@berdmann.de> for reporting the bug.
40
41 6. Made dump able to backup files larger than 2 GB. Note that
42 dump still doesn't cope with files larger than 4 GB.
43
44 7. Restore the real uid and gid before invoking an external
45 RSH program (big hole when dump or restore is suid root!).
46
47 8. Get the values for the default tape device and the location
48 of fstab file from the system headers. Thanks to
49 Andreas Dilger <adilger@turbolinux.com> for the patch.
50
51 9. Made dump -w|-W report all recognized filesystems
52 present in either /etc/fstab or /etc/dumpdates, and present
53 the list in the same order as in fstab file. Thanks
54 to Andreas Dilger <adilger@turbolinux.com> for the patch.
55
56 10. Made dump's -a (automatic end of tape detection) the
57 default. Specifying one of -B, -c, -d or -s options will
58 override the EOM detection. Thanks to Andreas Dilger
59 <adilger@turbolinux.com> for the patch.
60
61 11. Save the ext2 filesystem label into the dump volume label.
62 Specifying a volume label on the command line (-L option)
63 will override this feature. Thanks to Andreas Dilger
64 <adilger@turbolinux.com> for the patch.
65
66 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
67 =====================================================================
68
69 1. Fixed the signal handling in dump (which I broke in 0.4b17)
70 which was causing several strange dump failures (dump
71 hanged or segmentation faults etc).
72
73 2. Specified the default blocksize in the dump man page.
74
75 3. Changed two info messages of restore to be written on stdout
76 instead of stderr in order to leave stderr only for errors
77 or warnings. Thanks to Stephen Atwell
78 <satwell@urbana.css.mot.com> for the suggestion.
79
80 4. Corrected an off by one calculation which prevented
81 dumping a directory having a 1 character name length.
82 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
83 for reporting the bug.
84
85 5. Reinforce dump in order to allow only level 0 and no
86 -u option when dumping a subdirectory, and document
87 this in the man page. Thanks to Bernhard Erdmann
88 <bernhard.erdmann@gmx.de> for reporting the bug.
89
90 6. Fixed a small harmless bug in restore which caused
91 in some conditions hard links to be restored several
92 times (therefore generation some warning messages).
93 Thanks to Kenneth Porter <shiva@well.com> for
94 reporting the bug.
95
96 7. Updated the RPM spec file to the latest RedHat version,
97 providing FHS packaging and other cosmetic changes.
98 You will need RPM version >= 3.0.5 in order to build the RPM.
99
100 8. Updated the configure script to check for libtermcap
101 before checking for libreadline, because we need this
102 library in order to compile the readline support.
103
104 9. Made dump understand the LABEL= and UUID= notation
105 both in /etc/fstab and on the command line. Note that
106 you will need the /proc filesystem in order to use
107 these notations. Thanks to Erik Troan <ewt@redhat.com>
108 for providing the patch.
109
110 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
111 ===================================================================
112
113 1. Fixed a potential buffer overflow in restore. Thanks
114 to Stan Bubrouski <satan@fastdial.net> for reporting
115 the bug.
116
117 2. Fixed a readline-related bug which prevented
118 'cat DUMPFILE | restore -i -f -' from working. Thanks
119 to Charles Karney <karney@users.sourceforge.net>
120 for the bug report.
121
122 3. Changed a few "panic" into "exit", causing restore to
123 be more stable against some attacks (like the last one
124 reported on Bugtraq, although the last version of restore
125 was not vulnerable - just dumped core). Thanks to
126 Andreas Hasenack <andreas@conectiva.com.br> for reporting
127 the bugs to me.
128
129 4. Removed the suid-root bit on dump and restore in the
130 default build (and generated RPMs). It should be safer
131 now. Users who need the suid-root capabilities in order
132 to make network backups should read first the man page
133 of dump and enable the suid bit manually.
134
135 5. Added -ltermcap to the compile parameters for restore
136 when using readline support, in order to make the compile
137 process work on some systems (SuSE ?). Thanks to
138 Patrik Schindler <poc@pocnet.net> for reporting the bug.
139
140 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
141 ====================================================================
142
143 1. The -F script is called now *only* at the end of a tape,
144 not at the end of the dump. Thanks to Christian Haul
145 <haul@informatik.tu-darmstadt.de> for the bug report.
146
147 Normally, the device name and the current volume number
148 are passed on the command line of the script. However,
149 if you want the old style script (with no arguments
150 passed on the command line), you can enable it in
151 configure with the --enable-oldstylefscript.
152
153 2. Use posix signal handling to preserve dump functionality
154 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
155 the patch.
156
157 3. Made the exit code of restore in case of a 'C'ompare
158 command reflect the result of the compare. An exit status
159 of 0 means the dump archive is correct, 1 means tape errors,
160 2 means that some files were modified. Thanks to Kenneth Porter
161 <shiva@well.com> for the suggestion.
162
163 4. Made (finally) quotes work in interactive restore.
164
165 5. Small fixes in order to allow dump to compile with a
166 really old e2fsprogs version. Thanks to Ian Zimmerman
167 <itz@speakeasy.org> for the bug report.
168
169 6. Add GNU readline capabilities to interactive restore.
170 Use configure's --enable-readline flag to enable this feature.
171 Thanks to Patrik Schindler <poc@pocnet.net> for the
172 suggestion.
173
174 7. Do the compare on the fly in restore 'C' mode (this will
175 allow not to exhaust the available /tmp space when
176 ccmparing large files). Thanks to Kenneth Porter
177 <shiva@well.com> for the suggestion.
178
179 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
180 ====================================================================
181
182 1. Fixed some several duplicate 'DUMP: DUMP:' in the
183 output of dump.
184
185 2. Corrected the estimation of blocks to dump. Note that
186 this is still wrong for large directory inodes, due
187 to the size of a BSD directory entry on the tape
188 (estimation is based on the size of the ext2 directory,
189 which is smaller because it doesn't contain the null
190 character at the end).
191
192 3. Made dump report the total number of MB written to
193 tape at the end of run. Thanks to W. Reilly Cooley
194 <wcooley@nakedape.cc> for the patch.
195
196 4. Added the -X option to restore giving the possibility
197 to read the names of the files to be extracted/listed
198 from a text file (in addition of the command line).
199 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
200 patch.
201
202 5. Added the device name and the current volume number
203 as arguments to the end of tape script (dump -F option).
204
205 6. Made the multi-volume dump work again (after having
206 broken it in 0.4b15).
207
208 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
209 ===================================================================
210
211 1. Added a prompt command in interactive restore mode. Thanks
212 to Andreas Dilger <adilger@home.com> for the patch.
213
214 2. Fixed a buffer overflow problem in dump (caused by
215 not checking the size of the filesystem parameter).
216 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
217 reporting this on Bugtraq (and to several dump users
218 who forwarded me his mail).
219
220 3. Added the '-F script' option to dump in order to
221 launch a script at the end of each tape (to be used
222 with a tape changer, or to notify the sysadmin by
223 pager etc.).
224
225 4. Fixed a bug in restore compare code caused by the changes
226 I made in 0.4b14.
227
228 5. Fixed the treatment of options using the old BSD syntax
229 in both dump and restore.
230
231 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
232 =======================================================================
233
234 1. Fixed a bug in dump which may have caused invalid deleted
235 directories to be dumped out if they were deleted after the
236 mapping pass. This could occure on active filesystem and lead
237 to heap corruption (causing dump malfunction in many possible ways).
238 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
239 bug and submitting the patch.
240
241 2. Fixed the handling of the filesystem parameter in dump. A
242 '/mnt/disk' parameter caused the disk contents to be dumped,
243 but a '/mnt/disk/' parameter caused the mountpoint directory
244 to be dumped (generally an empty directory).
245
246 3. Improved the output of dump in order to tell which directory
247 it is currently dumping (when dumping a subtree).
248
249 4. Added the '-e' exclude inode option to dump. Thanks to
250 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
251
252 5. Added a REPORTING-BUGS file in order to provide a guide
253 on how to correctly report a bug in dump/restore.
254
255 6. Fixed a restore problem when restoring a hard link to an inode
256 having the immutable or append-only attribute set. Thanks to
257 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
258
259 7. Fixed a compatibility problem between dumps made on little
260 endian machines (the format was correct) and big endian
261 machines (the format was incorrect). This fix break the
262 compatibility with the older dumps made on big endian
263 machines (sparc, m86k, ppc etc). For the first time in
264 linux dump's history (I believe), the dumps made by *BSD,
265 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
266 so interchangeable. Thanks to Rob Cermak
267 <cermak@ahab.rutgers.edu> for submitting the bug and
268 helping me test the fix.
269
270 8. Fixed the way dump reports the remaining percent/time, if
271 the number of blocks actually dumped exceeds the estimated
272 number of blocks. Thanks to Jean-Paul van der Jagt
273 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
274
275 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
276 ======================================================================
277
278 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
279 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
280
281 2. Removed the requirement to build the RPM as root from the
282 spec file. Thanks to Christian Weisgerber
283 <naddy@mips.rhein-neckar.de> for submitting this.
284
285 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
286 causing dump to block when dumping really huge filesystems.
287 Many thanks to Patrik Schindler <poc@pocnet.net> for
288 helping me find this bug.
289
290 4. Fixed the treatment for an interrupt signal when dump access
291 the remote tape through RSH. Thanks to Christian Weisgerber
292 <naddy@mips.rhein-neckar.de> for providing the patch.
293
294 5. Fixed a bug which was causing dump/restore to display
295 garbage characters instead of the remote host name.
296
297 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
298 =====================================================================
299
300 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
301 <kukuk@suse.de> for submitting the patch.
302
303 2. Fix for the exit code when using the size estimate option of
304 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
305 submitting the patch.
306
307 3. Handle EINTR in atomical reads/writes in dump, which was causing
308 dump to fail on some systems. Thanks to Eric Jergensen
309 <eric@dvns.com> for reporting the bug and submitting the patch.
310
311 4. Handle more than 16 characters for the device names in dumpdates.
312 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
313 tracking down the problem and proposing the solution.
314
315 5. Fixed a bug in dump which prevented the creation of the
316 dumpdates file when doing a 0-level dump without already
317 having a dumpdates file. Thanks to Patrik Schindler
318 <poc@pocnet.net> for reporting the bug.
319
320 6. Changed the way dump 'S' flag reports the size estimate
321 from number of blocks into bytes (making it compatible
322 with the Solaris version, and simplifying things for
323 amanda users). Thanks to Jason L Tibbitts III
324 <tibbs@math.uh.edu> for reporting the bug.
325
326 7. Fixed a compatibility problem in linux/alpha dump tape format.
327 Now the linux/alpha dump are (again) compatible with the
328 other dump formats. But this breaks compatibility with
329 older dumps made on alpha. Thanks to Mike Tibor
330 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
331
332 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
333 ======================================================================
334
335 1. Added a '--enable-kerberos' to configure.
336
337 2. Added a 'S' option to dump which determines the amount of space
338 that is needed to perform the dump without actually doing it, similar
339 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
340 <cermak@ahab.rutgers.edu>.
341
342 3. Added a 'M' multi-volume option to dump and restore which enables
343 dumping to multiple files (useful when dumping to an ext2
344 partition to make several dump files in order to bypass the 2GB
345 file size limitation). The 'f' argument is treated as a prefix and
346 the output files will be named <prefix>001, <prefix>002 etc. With
347 the 'M' flag, restore automatically selects the right file without
348 asking to enter a new tape each time.
349
350 4. Fixed a memory leak which was causing dump to grow very big
351 (270MB when dumping a 10GB filesystem...). Thanks to Jason
352 Fearon <jasonf@netrider.org.au> for reporting the bug.
353
354 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
355 ======================================================================
356
357 1. Make configure test if the system glob routines support
358 extended syntax (ALTDIRFUNC). If not, use the internal glob
359 routines instead of system ones. Thanks to Bernhard Sadlowski
360 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
361 and helping me resolve this and other minor libc5 compiling
362 glitches.
363
364 2. Fix a problem when dumping a ext2fs with the 'filetype'
365 feature enabled. Thanks to Patrick J. LoPresti
366 <patl@cag.lcs.mit.edu> for reporting the bug and to
367 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
368
369 3. Made the nodump flag work on directories. A directory which
370 has the nodump flag gets never dumped, regardless of its
371 contents.
372
373 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
374 which allows dump on an active ext3 filesystem. However, this
375 is a "quick and dirty" patch which enables backup of an ext3
376 filesystem through the ext2 compatibility (by ignoring the
377 NEEDS_RECOVERY bit). The journal file is not recognized and
378 it is dumped (it should not).
379
380 5. Test the superblock compatibility flags when dumping, in order
381 to be sure that we know how to deal with specific features.
382
383 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
384 ====================================================================
385
386 1. Use lchown instead of chown, fixing a possible security problem
387 when restoring symlinks (a malicious user could use this
388 to deliberately corrupt the ownership of important system files).
389 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
390 this and providing the patch.
391
392 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
393 ====================================================================
394
395 1. Put dump sources under CVS, added Id tags in all files so
396 one can use 'ident' on binary files.
397
398 2. Added the dump/restore version in the usage text so one can
399 easily verify the version he is using.
400
401 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
402 a va_start/va_end problem on linux-ppc (always call va_start
403 va_end in pairs each time we use a vararg function).
404
405 4. Added again the DT_* constants because old libc does not
406 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
407 for submitting the bug report.
408
409 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
410 this should enable dumping big (huge) filesystems.
411
412 6. Added the RSH environment variable in order to be able to
413 use a rsh replacement like ssh when doing remote backups (and
414 bypass the security limitations of rcmd). Now you can do remote
415 backups without being root (or making dump setuid root).
416
417 7. Modified again the way dumpdates works. For incremental dumps,
418 we need to read dumpdates even if we are not using 'u' option.
419 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
420 this should work.
421
422 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
423 ===================================================================
424
425 1. Removed the 'k' flag from the restore 'about' text if kerberos
426 was not compiled in.
427
428 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
429 to them (fsetflags takes a char*, setflags an open fd!).
430
431 3. (f)setflags is called only if the flags aren't empty. If the
432 file is a special file, a warning is printed, because changing
433 flags implies opening the device. Normally, a special file
434 should not have any flag... (Debian bug #29775, patch provided
435 by Abhijit Dasgupta <abhijit@ans.net>).
436
437 4. Made possible to dump a file system not mentioned in /etc/fstab.
438 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
439
440 5. Changed the default behaviour to not create dumpdates
441 unless 'u' option is specified. Removed the old "debian-patch"
442 which provided the same thing. (Debian bug #38136, #33818).
443
444 6. Removed all those dump*announce, since they were getting old...
445
446 7. Added warning messages if dumpdates does not exist and
447 when an inferior level dump does not exist (except for a level 0
448 dump).
449
450 8. Debugged the glob calls in interactive mode: restore used a
451 dirent struct which was different from the /usr/include/dirent.h
452 one (this used to work, is it a glibc2 change?), so none of the
453 compat glob (which used /usr/include/dirent.h) or the system glob
454 worked. Restore use now the system dirent (and the system
455 DT_* constants), which are compatible with BSD ones.
456
457 9. Added a configure flag (--with-dumpdatespath) to specify
458 the location of dumpdates file. By default, it is
459 /etc/dumpdates.
460
461 10. Added the "AUTHOR" and "AVAILABILITY" sections and
462 included the current date/version in man pages.
463
464 11. Corrected the estimation of remaining time when
465 the operator doesn't change the tapes quickly enough. This
466 was an old bug, I thought I corrected it, and discovered
467 that in fact it was corrected in two different places, so
468 the results canceled each other...
469
470 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
471 ===================================================================
472
473 1. Integrated multiple patches from RedHat, Debian and SuSE:
474
475 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
476 isn't specified.
477 - several fixes in the man pages.
478 - update the default tape device to /dev/st0.
479 - many updates for Linux Alpha (byte ordering, size_t etc).
480 - buffer overruns.
481 - use environment variable for TMPDIR (instead of /tmp).
482 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
483 - workaround egcs bug (RedHat bugs #4281 and #2989).
484 - wire $(OPT) throughout Makefile's.
485
486 2. Upgrade the dump revision to 1, making possible to dump filesystems
487 made with e2fsprogs-1.15 or newer. Nothing seems to break...
488
489 3. Fix some compile warnings, prototype all functions.
490
491 4. Use glibc err/glob instead of internal compatibility
492 routines (only if available).
493
494 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
495 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
496
497 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
498 ======================================================================
499
500 1. Integrated the changes from FreeBSD-3.1-RELEASE
501 (mostly bug fixes, buffer overruns, dump has now an "automatic
502 tape length calculation" flag, dump/restore can use kerberos now
503 (this is NOT tested), use environment variables for TAPE and
504 RMT etc.).
505
506 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
507
508 3. Corrected a bug in restore when using 'C' option with multi-volumes
509 tapes (files splited accros two tapes give "size changed" errors
510 when comparing).
511
512 4. Corrected the long standing bug when dumping multiple tapes.
513 This works for me, needs further testing.
514
515 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
516 ====================================================================
517
518 1. Dump now runs correctly on kernels 2.1.x
519 Fix made by Gerald Peters <gapeters@worldnet.att.net>
520
521 Changes between versions 0.4b2 and 0.4b3
522 ========================================
523
524 1. Use realpath() if available
525
526 2. Report statistics
527
528 Changes between versions 0.4b1 and 0.4b2
529 ========================================
530
531 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
532 the patch)
533
534 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
535
536 3. Fixed configure.in to correctly handle --enable-rmt
537
538 Changes between versions 0.3 and 0.4b1
539 ======================================
540
541 1. Integrated the changes from 4.4BSD-Lite2
542
543 2. Integrated the patches from Debian and Red Hat
544
545 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
546
547 4. Changed dump to use the Ext2fs library to get block addresses. This
548 should solve the endianness problem on SparcLinux.
549
550 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
551 distribution's one) to use autoconf
552
553 6. Fixed a few minor bugs
554
555 Changes between versions 0.2e and 0.2f
556 ======================================
557
558 1. Added the creation of named pipes (fifos) in restore.
559
560 2. Added the -N flag in the restore manual page.
561
562 3. Added the file kernel.patch which contains the llseek() optimization
563 patch for 1.2.x kernels.
564
565 4. Fixed a bug in the restoration of symbolic links: owner and group were
566 not restored.
567
568 5. Integrated some changes from FreeBSD 2.2.
569
570 6. Added a call to ftruncate() after restoring each file to restore
571 correctly files ending by a hole.
572
573 Changes between versions 0.2d and 0.2e
574 ======================================
575
576 1. Fixed a bug in the "set owner/mode" process. Every file was restored
577 with owner = root (0) and group = root/wheel/whatever (0).
578
579 Changes between versions 0.2c and 0.2d
580 ======================================
581
582 1. Dump is now able to backup 2GB+ filesystems.
583
584 2. Dump and restore can now be linked as static binaries.
585
586 Changes between versions 0.2b and 0.2c
587 ======================================
588
589 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
590
591 Changes between versions 0.2a and 0.2b
592 ======================================
593
594 1. Really fixed the bug that I should have corrected in 0.2a.
595
596 2. Enabled optimization again.
597
598 Changes between versions 0.2 and 0.2a
599 =====================================
600
601 1. Disabled the optimization during compilation.
602
603 Changes between versions 0.1 and 0.2
604 ====================================
605
606 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
607 the fs_type field (actually, I modified the file fstab.c to make it
608 use the mntent functions).
609
610 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
611 made by dump should be readable by the BSD restore and backups made
612 by the BSD dump should be readable by restore. Unfortunately, this
613 means that the dump format has changed between version 0.1 and version
614 0.2 :-(
615
616 3. Dump is now able to backup a subtree, it is no longer limited to whole
617 filesystems like the BSD version.
618
619 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
620 than 2 GB.
621
622 Changes between versions 0.0 and 0.1
623 ====================================
624
625 1. Now create links rdump and rrestore during the `make install' step.
626
627 2. Linux port specific bugs added to the manual pages
628
629 3. Incorrect estimation of the number of tapes blocks fixed when doing
630 incremental backups.
631
632 4. Better ls-like format in restore in interactive mode.