]> git.wh0rd.org - dump.git/blob - CHANGES
9cae80d4c246bd2723ff34545900173ea8c81820
[dump.git] / CHANGES
1 $Id: CHANGES,v 1.123 2001/07/18 13:12:33 stelian Exp $
2
3 Changes between versions 0.4b22 and 0.4b23 (released ????????????)
4 ==================================================================
5
6 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
7 Marcus Meissner (Caldera International Security Dept.).
8
9 2. Implement the Sun rmt extended protocol. Patch submitted
10 by Ian Gordon <iangordon@users.sourceforge.net>.
11
12 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
13 the linux kernel header. This ensures that dump always has the
14 latest version of this file. Patch submitted by
15 Andreas Dilger <adilger@turbolinux.com>.
16
17 4. Report any filesystem present in either /etc/fstab with a
18 positive passno or /etc/dumpdates in dump -w output.
19 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
20
21 5. Fixed the looping problem in dump introduced in the
22 previous version.
23
24 6. Changed the -B option of dump to limit the size of
25 _compressed_ output per volume if compression is on.
26 Patch contributed by Helmut Jarausch
27 <jarausch@igpm.rwth-aachen.de>.
28
29 7. Fixed a bug in reading the operator typed file/tape path for
30 the next volume in restore.
31
32 8. Implemented a "-F script" option to restore which permits the
33 user to specify a script which will be launched at the
34 beginning of each tape, useful for automatic programming of
35 tape changers for example. See the restore man page for the
36 script parameters and return codes.
37
38 9. Small fix for the QFA routines provided by Uwe Gohlke
39 <uwe@ugsoft.de>, and some recommendations for QFA uses in
40 the man pages.
41
42 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
43 ==================================================================
44
45 1. Made dump report the number of blocks dumped per volume.
46 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
47
48 2. Fix a bug in dump where waiting too much at the 'change volumes'
49 question caused the volume to be rewritten. Thanks to
50 Graham Reed <greed@users.sourceforge.net> for reporting the
51 bug and providing a patch.
52
53 3. Added a compression option to dump, contributed by Jerry
54 Peters <gapeters@worldnet.att.net>.
55
56 WARNING: the compression works only on files/pipes and
57 drives supporting variable block size.
58
59 WARNING2: tapes created using the compression feature are
60 incompatible with the standard dump tape format, and a
61 version of restore >= 0.4b22 will be needed for extraction.
62
63 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
64 architectures. Thanks to Paul Slootman <paul@debian.org> for
65 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
66 it upstream.
67
68 5. Many cleanups (CPP defines, const char warnings, check of
69 ext2fs COMPAT flags, time_t cleanups, added libext2 version
70 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
71
72 6. Made --prefix option work in configure. All the install path
73 are now based on the configure parameters.
74
75 7. Added the Quick File Access mode in dump/restore, contributed
76 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
77 a file tape position for each inode, and this file is used by
78 restore (if called with parameter Q and the filename)
79 to directly position the tape at the file restore is currently
80 working on. This saves hours when restoring single files from
81 large backups, saves the tapes and the drive's head. Use
82 --enable-qfa option of configure to compile in the QFA support.
83
84 8. Added the possibility to dump several files and directories
85 in a single invocation of dump. Thanks to Uwe Gohlke
86 <uwe@ugsoft.de> for implementing this option.
87
88 9. Fixed the dumping and restoring of files with holes
89 (files ending with holes didn't get dumped or restored
90 correctly).
91
92 10. Fixed a socket descriptor leak in dump, which leaved opened
93 3 file descriptors per dump process (and there is one dump
94 process per tape).
95
96 11. Fixed dump large file system support, by explicit use of
97 open64/lseek64/etc functions (explicit use needed because
98 e2fsprogs libraries don't behave well when compiled with
99 FILE_OFFSET_BITS=64).
100
101 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
102 ======================================================================
103
104 1. Fixed some bugs in the dump -w|-W logic introduced by
105 the previous version. Thanks to Andreas Dilger
106 <adilger@turbolinux.com> for his help on this one.
107
108 2. Fixed again a compilation problem when using old e2fs
109 headers (filesystem label related). Thanks to many users
110 who reported this stupid error.
111
112 3. Fixed a build problem on old lib5 based systems dealing with
113 _PATH_MNTTAB being undefined. Thanks to John Adams
114 <johna@onevista.com> for reporting the bug.
115
116 4. Improved the error detection in restore when creating
117 the temporary files in TMPDIR. Now restore will corectly
118 report a 'no space left on device' error instead of
119 strange errors which could imply an archive corruption.
120 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
121 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
122
123 5. Added the throughput information in dump's progression
124 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
125 for the patch.
126
127 6. Use libext2fs's inode scanning routines, which are particularly
128 robust when dealing with errors (especially when having some
129 bad blocks on the filesystem). Many thanks to Colin
130 <colin@horizon.com> for reporting the bug, investigating
131 the issues, finding a workaround, writing the patch and
132 fully testing it... (of course, if this patch will break
133 anything, he is to blame for :-)).
134
135 7. Made dump and restore LFS aware. Dump can dump a filesystem
136 containing large files, generate a large file on output and
137 restore can restore them. This feature is not enabled by
138 default, use --enable-largefile option of configure to enable
139 it (you need to have a LFS aware glibc though). Thanks to
140 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
141 and to Theodore T'so <tytso@valinux.com> for his always
142 useful thoughts.
143
144 8. Made dump ask upon a tape write error if it should rewrite
145 the current volume (assume this is a bad tape) or if it should
146 assume an end-of-tape condition (useful for tape drives which
147 behaves badly at the end of the tape). Thanks to Andreas
148 Dilger <adilger@turbolinux.com> for the suggestion.
149
150 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
151 =======================================================================
152
153 1. Fixed a small compilation problem due to a change
154 in the definintion of the struct sigaction in
155 glibc 2.0 / libc5. Thanks to Gunther Schlegel
156 <schlegel@riege.de> for reporting the bug and to
157 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
158 a fix.
159
160 2. Modified the label and uuid analysis in order to be
161 self-contained (independant of kernel/libc headers). This
162 should fix the compile with older kernel/libc headers and
163 will preserve the functionality. Thanks to Bernhard Erdmann
164 <bernhard.erdmann@gmx.de> for reporting the bug.
165
166 3. The 'exclude inode' option, if applied to a directory
167 inode, excludes now correctly all the leaves of this
168 directory. Thanks to John R. Dennison
169 <gerdesas@users.sourceforge.net> for reporting the bug.
170
171 4. Fixed the '-e' option to disable the possibility
172 to exclude the root inode (which causes the dumps to
173 be unreadable by restore). Prevented array overflow
174 when multiple -e options are used.
175
176 5. Fixed dump to correctly interpret a filesystem argument
177 which is a mountpoint and it is not an absolute path
178 (as specified in the fstab). Thanks to Bernhard R. Erdmann
179 <be@berdmann.de> for reporting the bug.
180
181 6. Made dump able to backup files larger than 2 GB. Note that
182 dump still doesn't cope with files larger than 4 GB.
183
184 7. Restore the real uid and gid before invoking an external
185 RSH program (big hole when dump or restore is suid root!).
186
187 8. Get the values for the default tape device and the location
188 of fstab file from the system headers. Thanks to
189 Andreas Dilger <adilger@turbolinux.com> for the patch.
190
191 9. Made dump -w|-W report all recognized filesystems
192 present in either /etc/fstab or /etc/dumpdates, and present
193 the list in the same order as in fstab file. Thanks
194 to Andreas Dilger <adilger@turbolinux.com> for the patch.
195
196 10. Made dump's -a (automatic end of tape detection) the
197 default. Specifying one of -B, -c, -d or -s options will
198 override the EOM detection. Thanks to Andreas Dilger
199 <adilger@turbolinux.com> for the patch.
200
201 11. Save the ext2 filesystem label into the dump volume label.
202 Specifying a volume label on the command line (-L option)
203 will override this feature. Thanks to Andreas Dilger
204 <adilger@turbolinux.com> for the patch.
205
206 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
207 =====================================================================
208
209 1. Fixed the signal handling in dump (which I broke in 0.4b17)
210 which was causing several strange dump failures (dump
211 hanged or segmentation faults etc).
212
213 2. Specified the default blocksize in the dump man page.
214
215 3. Changed two info messages of restore to be written on stdout
216 instead of stderr in order to leave stderr only for errors
217 or warnings. Thanks to Stephen Atwell
218 <satwell@urbana.css.mot.com> for the suggestion.
219
220 4. Corrected an off by one calculation which prevented
221 dumping a directory having a 1 character name length.
222 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
223 for reporting the bug.
224
225 5. Reinforce dump in order to allow only level 0 and no
226 -u option when dumping a subdirectory, and document
227 this in the man page. Thanks to Bernhard Erdmann
228 <bernhard.erdmann@gmx.de> for reporting the bug.
229
230 6. Fixed a small harmless bug in restore which caused
231 in some conditions hard links to be restored several
232 times (therefore generation some warning messages).
233 Thanks to Kenneth Porter <shiva@well.com> for
234 reporting the bug.
235
236 7. Updated the RPM spec file to the latest RedHat version,
237 providing FHS packaging and other cosmetic changes.
238 You will need RPM version >= 3.0.5 in order to build the RPM.
239
240 8. Updated the configure script to check for libtermcap
241 before checking for libreadline, because we need this
242 library in order to compile the readline support.
243
244 9. Made dump understand the LABEL= and UUID= notation
245 both in /etc/fstab and on the command line. Note that
246 you will need the /proc filesystem in order to use
247 these notations. Thanks to Erik Troan <ewt@redhat.com>
248 for providing the patch.
249
250 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
251 ===================================================================
252
253 1. Fixed a potential buffer overflow in restore. Thanks
254 to Stan Bubrouski <satan@fastdial.net> for reporting
255 the bug.
256
257 2. Fixed a readline-related bug which prevented
258 'cat DUMPFILE | restore -i -f -' from working. Thanks
259 to Charles Karney <karney@users.sourceforge.net>
260 for the bug report.
261
262 3. Changed a few "panic" into "exit", causing restore to
263 be more stable against some attacks (like the last one
264 reported on Bugtraq, although the last version of restore
265 was not vulnerable - just dumped core). Thanks to
266 Andreas Hasenack <andreas@conectiva.com.br> for reporting
267 the bugs to me.
268
269 4. Removed the suid-root bit on dump and restore in the
270 default build (and generated RPMs). It should be safer
271 now. Users who need the suid-root capabilities in order
272 to make network backups should read first the man page
273 of dump and enable the suid bit manually.
274
275 5. Added -ltermcap to the compile parameters for restore
276 when using readline support, in order to make the compile
277 process work on some systems (SuSE ?). Thanks to
278 Patrik Schindler <poc@pocnet.net> for reporting the bug.
279
280 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
281 ====================================================================
282
283 1. The -F script is called now *only* at the end of a tape,
284 not at the end of the dump. Thanks to Christian Haul
285 <haul@informatik.tu-darmstadt.de> for the bug report.
286
287 Normally, the device name and the current volume number
288 are passed on the command line of the script. However,
289 if you want the old style script (with no arguments
290 passed on the command line), you can enable it in
291 configure with the --enable-oldstylefscript.
292
293 2. Use posix signal handling to preserve dump functionality
294 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
295 the patch.
296
297 3. Made the exit code of restore in case of a 'C'ompare
298 command reflect the result of the compare. An exit status
299 of 0 means the dump archive is correct, 1 means tape errors,
300 2 means that some files were modified. Thanks to Kenneth Porter
301 <shiva@well.com> for the suggestion.
302
303 4. Made (finally) quotes work in interactive restore.
304
305 5. Small fixes in order to allow dump to compile with a
306 really old e2fsprogs version. Thanks to Ian Zimmerman
307 <itz@speakeasy.org> for the bug report.
308
309 6. Add GNU readline capabilities to interactive restore.
310 Use configure's --enable-readline flag to enable this feature.
311 Thanks to Patrik Schindler <poc@pocnet.net> for the
312 suggestion.
313
314 7. Do the compare on the fly in restore 'C' mode (this will
315 allow not to exhaust the available /tmp space when
316 ccmparing large files). Thanks to Kenneth Porter
317 <shiva@well.com> for the suggestion.
318
319 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
320 ====================================================================
321
322 1. Fixed some several duplicate 'DUMP: DUMP:' in the
323 output of dump.
324
325 2. Corrected the estimation of blocks to dump. Note that
326 this is still wrong for large directory inodes, due
327 to the size of a BSD directory entry on the tape
328 (estimation is based on the size of the ext2 directory,
329 which is smaller because it doesn't contain the null
330 character at the end).
331
332 3. Made dump report the total number of MB written to
333 tape at the end of run. Thanks to W. Reilly Cooley
334 <wcooley@nakedape.cc> for the patch.
335
336 4. Added the -X option to restore giving the possibility
337 to read the names of the files to be extracted/listed
338 from a text file (in addition of the command line).
339 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
340 patch.
341
342 5. Added the device name and the current volume number
343 as arguments to the end of tape script (dump -F option).
344
345 6. Made the multi-volume dump work again (after having
346 broken it in 0.4b15).
347
348 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
349 ===================================================================
350
351 1. Added a prompt command in interactive restore mode. Thanks
352 to Andreas Dilger <adilger@home.com> for the patch.
353
354 2. Fixed a buffer overflow problem in dump (caused by
355 not checking the size of the filesystem parameter).
356 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
357 reporting this on Bugtraq (and to several dump users
358 who forwarded me his mail).
359
360 3. Added the '-F script' option to dump in order to
361 launch a script at the end of each tape (to be used
362 with a tape changer, or to notify the sysadmin by
363 pager etc.).
364
365 4. Fixed a bug in restore compare code caused by the changes
366 I made in 0.4b14.
367
368 5. Fixed the treatment of options using the old BSD syntax
369 in both dump and restore.
370
371 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
372 =======================================================================
373
374 1. Fixed a bug in dump which may have caused invalid deleted
375 directories to be dumped out if they were deleted after the
376 mapping pass. This could occure on active filesystem and lead
377 to heap corruption (causing dump malfunction in many possible ways).
378 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
379 bug and submitting the patch.
380
381 2. Fixed the handling of the filesystem parameter in dump. A
382 '/mnt/disk' parameter caused the disk contents to be dumped,
383 but a '/mnt/disk/' parameter caused the mountpoint directory
384 to be dumped (generally an empty directory).
385
386 3. Improved the output of dump in order to tell which directory
387 it is currently dumping (when dumping a subtree).
388
389 4. Added the '-e' exclude inode option to dump. Thanks to
390 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
391
392 5. Added a REPORTING-BUGS file in order to provide a guide
393 on how to correctly report a bug in dump/restore.
394
395 6. Fixed a restore problem when restoring a hard link to an inode
396 having the immutable or append-only attribute set. Thanks to
397 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
398
399 7. Fixed a compatibility problem between dumps made on little
400 endian machines (the format was correct) and big endian
401 machines (the format was incorrect). This fix break the
402 compatibility with the older dumps made on big endian
403 machines (sparc, m86k, ppc etc). For the first time in
404 linux dump's history (I believe), the dumps made by *BSD,
405 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
406 so interchangeable. Thanks to Rob Cermak
407 <cermak@ahab.rutgers.edu> for submitting the bug and
408 helping me test the fix.
409
410 8. Fixed the way dump reports the remaining percent/time, if
411 the number of blocks actually dumped exceeds the estimated
412 number of blocks. Thanks to Jean-Paul van der Jagt
413 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
414
415 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
416 ======================================================================
417
418 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
419 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
420
421 2. Removed the requirement to build the RPM as root from the
422 spec file. Thanks to Christian Weisgerber
423 <naddy@mips.rhein-neckar.de> for submitting this.
424
425 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
426 causing dump to block when dumping really huge filesystems.
427 Many thanks to Patrik Schindler <poc@pocnet.net> for
428 helping me find this bug.
429
430 4. Fixed the treatment for an interrupt signal when dump access
431 the remote tape through RSH. Thanks to Christian Weisgerber
432 <naddy@mips.rhein-neckar.de> for providing the patch.
433
434 5. Fixed a bug which was causing dump/restore to display
435 garbage characters instead of the remote host name.
436
437 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
438 =====================================================================
439
440 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
441 <kukuk@suse.de> for submitting the patch.
442
443 2. Fix for the exit code when using the size estimate option of
444 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
445 submitting the patch.
446
447 3. Handle EINTR in atomical reads/writes in dump, which was causing
448 dump to fail on some systems. Thanks to Eric Jergensen
449 <eric@dvns.com> for reporting the bug and submitting the patch.
450
451 4. Handle more than 16 characters for the device names in dumpdates.
452 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
453 tracking down the problem and proposing the solution.
454
455 5. Fixed a bug in dump which prevented the creation of the
456 dumpdates file when doing a 0-level dump without already
457 having a dumpdates file. Thanks to Patrik Schindler
458 <poc@pocnet.net> for reporting the bug.
459
460 6. Changed the way dump 'S' flag reports the size estimate
461 from number of blocks into bytes (making it compatible
462 with the Solaris version, and simplifying things for
463 amanda users). Thanks to Jason L Tibbitts III
464 <tibbs@math.uh.edu> for reporting the bug.
465
466 7. Fixed a compatibility problem in linux/alpha dump tape format.
467 Now the linux/alpha dump are (again) compatible with the
468 other dump formats. But this breaks compatibility with
469 older dumps made on alpha. Thanks to Mike Tibor
470 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
471
472 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
473 ======================================================================
474
475 1. Added a '--enable-kerberos' to configure.
476
477 2. Added a 'S' option to dump which determines the amount of space
478 that is needed to perform the dump without actually doing it, similar
479 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
480 <cermak@ahab.rutgers.edu>.
481
482 3. Added a 'M' multi-volume option to dump and restore which enables
483 dumping to multiple files (useful when dumping to an ext2
484 partition to make several dump files in order to bypass the 2GB
485 file size limitation). The 'f' argument is treated as a prefix and
486 the output files will be named <prefix>001, <prefix>002 etc. With
487 the 'M' flag, restore automatically selects the right file without
488 asking to enter a new tape each time.
489
490 4. Fixed a memory leak which was causing dump to grow very big
491 (270MB when dumping a 10GB filesystem...). Thanks to Jason
492 Fearon <jasonf@netrider.org.au> for reporting the bug.
493
494 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
495 ======================================================================
496
497 1. Make configure test if the system glob routines support
498 extended syntax (ALTDIRFUNC). If not, use the internal glob
499 routines instead of system ones. Thanks to Bernhard Sadlowski
500 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
501 and helping me resolve this and other minor libc5 compiling
502 glitches.
503
504 2. Fix a problem when dumping a ext2fs with the 'filetype'
505 feature enabled. Thanks to Patrick J. LoPresti
506 <patl@cag.lcs.mit.edu> for reporting the bug and to
507 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
508
509 3. Made the nodump flag work on directories. A directory which
510 has the nodump flag gets never dumped, regardless of its
511 contents.
512
513 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
514 which allows dump on an active ext3 filesystem. However, this
515 is a "quick and dirty" patch which enables backup of an ext3
516 filesystem through the ext2 compatibility (by ignoring the
517 NEEDS_RECOVERY bit). The journal file is not recognized and
518 it is dumped (it should not).
519
520 5. Test the superblock compatibility flags when dumping, in order
521 to be sure that we know how to deal with specific features.
522
523 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
524 ====================================================================
525
526 1. Use lchown instead of chown, fixing a possible security problem
527 when restoring symlinks (a malicious user could use this
528 to deliberately corrupt the ownership of important system files).
529 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
530 this and providing the patch.
531
532 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
533 ====================================================================
534
535 1. Put dump sources under CVS, added Id tags in all files so
536 one can use 'ident' on binary files.
537
538 2. Added the dump/restore version in the usage text so one can
539 easily verify the version he is using.
540
541 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
542 a va_start/va_end problem on linux-ppc (always call va_start
543 va_end in pairs each time we use a vararg function).
544
545 4. Added again the DT_* constants because old libc does not
546 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
547 for submitting the bug report.
548
549 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
550 this should enable dumping big (huge) filesystems.
551
552 6. Added the RSH environment variable in order to be able to
553 use a rsh replacement like ssh when doing remote backups (and
554 bypass the security limitations of rcmd). Now you can do remote
555 backups without being root (or making dump setuid root).
556
557 7. Modified again the way dumpdates works. For incremental dumps,
558 we need to read dumpdates even if we are not using 'u' option.
559 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
560 this should work.
561
562 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
563 ===================================================================
564
565 1. Removed the 'k' flag from the restore 'about' text if kerberos
566 was not compiled in.
567
568 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
569 to them (fsetflags takes a char*, setflags an open fd!).
570
571 3. (f)setflags is called only if the flags aren't empty. If the
572 file is a special file, a warning is printed, because changing
573 flags implies opening the device. Normally, a special file
574 should not have any flag... (Debian bug #29775, patch provided
575 by Abhijit Dasgupta <abhijit@ans.net>).
576
577 4. Made possible to dump a file system not mentioned in /etc/fstab.
578 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
579
580 5. Changed the default behaviour to not create dumpdates
581 unless 'u' option is specified. Removed the old "debian-patch"
582 which provided the same thing. (Debian bug #38136, #33818).
583
584 6. Removed all those dump*announce, since they were getting old...
585
586 7. Added warning messages if dumpdates does not exist and
587 when an inferior level dump does not exist (except for a level 0
588 dump).
589
590 8. Debugged the glob calls in interactive mode: restore used a
591 dirent struct which was different from the /usr/include/dirent.h
592 one (this used to work, is it a glibc2 change?), so none of the
593 compat glob (which used /usr/include/dirent.h) or the system glob
594 worked. Restore use now the system dirent (and the system
595 DT_* constants), which are compatible with BSD ones.
596
597 9. Added a configure flag (--with-dumpdatespath) to specify
598 the location of dumpdates file. By default, it is
599 /etc/dumpdates.
600
601 10. Added the "AUTHOR" and "AVAILABILITY" sections and
602 included the current date/version in man pages.
603
604 11. Corrected the estimation of remaining time when
605 the operator doesn't change the tapes quickly enough. This
606 was an old bug, I thought I corrected it, and discovered
607 that in fact it was corrected in two different places, so
608 the results canceled each other...
609
610 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
611 ===================================================================
612
613 1. Integrated multiple patches from RedHat, Debian and SuSE:
614
615 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
616 isn't specified.
617 - several fixes in the man pages.
618 - update the default tape device to /dev/st0.
619 - many updates for Linux Alpha (byte ordering, size_t etc).
620 - buffer overruns.
621 - use environment variable for TMPDIR (instead of /tmp).
622 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
623 - workaround egcs bug (RedHat bugs #4281 and #2989).
624 - wire $(OPT) throughout Makefile's.
625
626 2. Upgrade the dump revision to 1, making possible to dump filesystems
627 made with e2fsprogs-1.15 or newer. Nothing seems to break...
628
629 3. Fix some compile warnings, prototype all functions.
630
631 4. Use glibc err/glob instead of internal compatibility
632 routines (only if available).
633
634 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
635 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
636
637 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
638 ======================================================================
639
640 1. Integrated the changes from FreeBSD-3.1-RELEASE
641 (mostly bug fixes, buffer overruns, dump has now an "automatic
642 tape length calculation" flag, dump/restore can use kerberos now
643 (this is NOT tested), use environment variables for TAPE and
644 RMT etc.).
645
646 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
647
648 3. Corrected a bug in restore when using 'C' option with multi-volumes
649 tapes (files splited accros two tapes give "size changed" errors
650 when comparing).
651
652 4. Corrected the long standing bug when dumping multiple tapes.
653 This works for me, needs further testing.
654
655 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
656 ====================================================================
657
658 1. Dump now runs correctly on kernels 2.1.x
659 Fix made by Gerald Peters <gapeters@worldnet.att.net>
660
661 Changes between versions 0.4b2 and 0.4b3
662 ========================================
663
664 1. Use realpath() if available
665
666 2. Report statistics
667
668 Changes between versions 0.4b1 and 0.4b2
669 ========================================
670
671 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
672 the patch)
673
674 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
675
676 3. Fixed configure.in to correctly handle --enable-rmt
677
678 Changes between versions 0.3 and 0.4b1
679 ======================================
680
681 1. Integrated the changes from 4.4BSD-Lite2
682
683 2. Integrated the patches from Debian and Red Hat
684
685 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
686
687 4. Changed dump to use the Ext2fs library to get block addresses. This
688 should solve the endianness problem on SparcLinux.
689
690 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
691 distribution's one) to use autoconf
692
693 6. Fixed a few minor bugs
694
695 Changes between versions 0.2e and 0.2f
696 ======================================
697
698 1. Added the creation of named pipes (fifos) in restore.
699
700 2. Added the -N flag in the restore manual page.
701
702 3. Added the file kernel.patch which contains the llseek() optimization
703 patch for 1.2.x kernels.
704
705 4. Fixed a bug in the restoration of symbolic links: owner and group were
706 not restored.
707
708 5. Integrated some changes from FreeBSD 2.2.
709
710 6. Added a call to ftruncate() after restoring each file to restore
711 correctly files ending by a hole.
712
713 Changes between versions 0.2d and 0.2e
714 ======================================
715
716 1. Fixed a bug in the "set owner/mode" process. Every file was restored
717 with owner = root (0) and group = root/wheel/whatever (0).
718
719 Changes between versions 0.2c and 0.2d
720 ======================================
721
722 1. Dump is now able to backup 2GB+ filesystems.
723
724 2. Dump and restore can now be linked as static binaries.
725
726 Changes between versions 0.2b and 0.2c
727 ======================================
728
729 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
730
731 Changes between versions 0.2a and 0.2b
732 ======================================
733
734 1. Really fixed the bug that I should have corrected in 0.2a.
735
736 2. Enabled optimization again.
737
738 Changes between versions 0.2 and 0.2a
739 =====================================
740
741 1. Disabled the optimization during compilation.
742
743 Changes between versions 0.1 and 0.2
744 ====================================
745
746 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
747 the fs_type field (actually, I modified the file fstab.c to make it
748 use the mntent functions).
749
750 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
751 made by dump should be readable by the BSD restore and backups made
752 by the BSD dump should be readable by restore. Unfortunately, this
753 means that the dump format has changed between version 0.1 and version
754 0.2 :-(
755
756 3. Dump is now able to backup a subtree, it is no longer limited to whole
757 filesystems like the BSD version.
758
759 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
760 than 2 GB.
761
762 Changes between versions 0.0 and 0.1
763 ====================================
764
765 1. Now create links rdump and rrestore during the `make install' step.
766
767 2. Linux port specific bugs added to the manual pages
768
769 3. Incorrect estimation of the number of tapes blocks fixed when doing
770 incremental backups.
771
772 4. Better ls-like format in restore in interactive mode.