]> git.wh0rd.org - dump.git/blob - CHANGES
Discard newline when reading the tape path entered by the operator in restore.
[dump.git] / CHANGES
1 $Id: CHANGES,v 1.121 2001/07/18 11:46:31 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 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
33 ==================================================================
34
35 1. Made dump report the number of blocks dumped per volume.
36 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
37
38 2. Fix a bug in dump where waiting too much at the 'change volumes'
39 question caused the volume to be rewritten. Thanks to
40 Graham Reed <greed@users.sourceforge.net> for reporting the
41 bug and providing a patch.
42
43 3. Added a compression option to dump, contributed by Jerry
44 Peters <gapeters@worldnet.att.net>.
45
46 WARNING: the compression works only on files/pipes and
47 drives supporting variable block size.
48
49 WARNING2: tapes created using the compression feature are
50 incompatible with the standard dump tape format, and a
51 version of restore >= 0.4b22 will be needed for extraction.
52
53 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
54 architectures. Thanks to Paul Slootman <paul@debian.org> for
55 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
56 it upstream.
57
58 5. Many cleanups (CPP defines, const char warnings, check of
59 ext2fs COMPAT flags, time_t cleanups, added libext2 version
60 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
61
62 6. Made --prefix option work in configure. All the install path
63 are now based on the configure parameters.
64
65 7. Added the Quick File Access mode in dump/restore, contributed
66 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
67 a file tape position for each inode, and this file is used by
68 restore (if called with parameter Q and the filename)
69 to directly position the tape at the file restore is currently
70 working on. This saves hours when restoring single files from
71 large backups, saves the tapes and the drive's head. Use
72 --enable-qfa option of configure to compile in the QFA support.
73
74 8. Added the possibility to dump several files and directories
75 in a single invocation of dump. Thanks to Uwe Gohlke
76 <uwe@ugsoft.de> for implementing this option.
77
78 9. Fixed the dumping and restoring of files with holes
79 (files ending with holes didn't get dumped or restored
80 correctly).
81
82 10. Fixed a socket descriptor leak in dump, which leaved opened
83 3 file descriptors per dump process (and there is one dump
84 process per tape).
85
86 11. Fixed dump large file system support, by explicit use of
87 open64/lseek64/etc functions (explicit use needed because
88 e2fsprogs libraries don't behave well when compiled with
89 FILE_OFFSET_BITS=64).
90
91 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
92 ======================================================================
93
94 1. Fixed some bugs in the dump -w|-W logic introduced by
95 the previous version. Thanks to Andreas Dilger
96 <adilger@turbolinux.com> for his help on this one.
97
98 2. Fixed again a compilation problem when using old e2fs
99 headers (filesystem label related). Thanks to many users
100 who reported this stupid error.
101
102 3. Fixed a build problem on old lib5 based systems dealing with
103 _PATH_MNTTAB being undefined. Thanks to John Adams
104 <johna@onevista.com> for reporting the bug.
105
106 4. Improved the error detection in restore when creating
107 the temporary files in TMPDIR. Now restore will corectly
108 report a 'no space left on device' error instead of
109 strange errors which could imply an archive corruption.
110 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
111 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
112
113 5. Added the throughput information in dump's progression
114 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
115 for the patch.
116
117 6. Use libext2fs's inode scanning routines, which are particularly
118 robust when dealing with errors (especially when having some
119 bad blocks on the filesystem). Many thanks to Colin
120 <colin@horizon.com> for reporting the bug, investigating
121 the issues, finding a workaround, writing the patch and
122 fully testing it... (of course, if this patch will break
123 anything, he is to blame for :-)).
124
125 7. Made dump and restore LFS aware. Dump can dump a filesystem
126 containing large files, generate a large file on output and
127 restore can restore them. This feature is not enabled by
128 default, use --enable-largefile option of configure to enable
129 it (you need to have a LFS aware glibc though). Thanks to
130 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
131 and to Theodore T'so <tytso@valinux.com> for his always
132 useful thoughts.
133
134 8. Made dump ask upon a tape write error if it should rewrite
135 the current volume (assume this is a bad tape) or if it should
136 assume an end-of-tape condition (useful for tape drives which
137 behaves badly at the end of the tape). Thanks to Andreas
138 Dilger <adilger@turbolinux.com> for the suggestion.
139
140 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
141 =======================================================================
142
143 1. Fixed a small compilation problem due to a change
144 in the definintion of the struct sigaction in
145 glibc 2.0 / libc5. Thanks to Gunther Schlegel
146 <schlegel@riege.de> for reporting the bug and to
147 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
148 a fix.
149
150 2. Modified the label and uuid analysis in order to be
151 self-contained (independant of kernel/libc headers). This
152 should fix the compile with older kernel/libc headers and
153 will preserve the functionality. Thanks to Bernhard Erdmann
154 <bernhard.erdmann@gmx.de> for reporting the bug.
155
156 3. The 'exclude inode' option, if applied to a directory
157 inode, excludes now correctly all the leaves of this
158 directory. Thanks to John R. Dennison
159 <gerdesas@users.sourceforge.net> for reporting the bug.
160
161 4. Fixed the '-e' option to disable the possibility
162 to exclude the root inode (which causes the dumps to
163 be unreadable by restore). Prevented array overflow
164 when multiple -e options are used.
165
166 5. Fixed dump to correctly interpret a filesystem argument
167 which is a mountpoint and it is not an absolute path
168 (as specified in the fstab). Thanks to Bernhard R. Erdmann
169 <be@berdmann.de> for reporting the bug.
170
171 6. Made dump able to backup files larger than 2 GB. Note that
172 dump still doesn't cope with files larger than 4 GB.
173
174 7. Restore the real uid and gid before invoking an external
175 RSH program (big hole when dump or restore is suid root!).
176
177 8. Get the values for the default tape device and the location
178 of fstab file from the system headers. Thanks to
179 Andreas Dilger <adilger@turbolinux.com> for the patch.
180
181 9. Made dump -w|-W report all recognized filesystems
182 present in either /etc/fstab or /etc/dumpdates, and present
183 the list in the same order as in fstab file. Thanks
184 to Andreas Dilger <adilger@turbolinux.com> for the patch.
185
186 10. Made dump's -a (automatic end of tape detection) the
187 default. Specifying one of -B, -c, -d or -s options will
188 override the EOM detection. Thanks to Andreas Dilger
189 <adilger@turbolinux.com> for the patch.
190
191 11. Save the ext2 filesystem label into the dump volume label.
192 Specifying a volume label on the command line (-L option)
193 will override this feature. Thanks to Andreas Dilger
194 <adilger@turbolinux.com> for the patch.
195
196 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
197 =====================================================================
198
199 1. Fixed the signal handling in dump (which I broke in 0.4b17)
200 which was causing several strange dump failures (dump
201 hanged or segmentation faults etc).
202
203 2. Specified the default blocksize in the dump man page.
204
205 3. Changed two info messages of restore to be written on stdout
206 instead of stderr in order to leave stderr only for errors
207 or warnings. Thanks to Stephen Atwell
208 <satwell@urbana.css.mot.com> for the suggestion.
209
210 4. Corrected an off by one calculation which prevented
211 dumping a directory having a 1 character name length.
212 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
213 for reporting the bug.
214
215 5. Reinforce dump in order to allow only level 0 and no
216 -u option when dumping a subdirectory, and document
217 this in the man page. Thanks to Bernhard Erdmann
218 <bernhard.erdmann@gmx.de> for reporting the bug.
219
220 6. Fixed a small harmless bug in restore which caused
221 in some conditions hard links to be restored several
222 times (therefore generation some warning messages).
223 Thanks to Kenneth Porter <shiva@well.com> for
224 reporting the bug.
225
226 7. Updated the RPM spec file to the latest RedHat version,
227 providing FHS packaging and other cosmetic changes.
228 You will need RPM version >= 3.0.5 in order to build the RPM.
229
230 8. Updated the configure script to check for libtermcap
231 before checking for libreadline, because we need this
232 library in order to compile the readline support.
233
234 9. Made dump understand the LABEL= and UUID= notation
235 both in /etc/fstab and on the command line. Note that
236 you will need the /proc filesystem in order to use
237 these notations. Thanks to Erik Troan <ewt@redhat.com>
238 for providing the patch.
239
240 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
241 ===================================================================
242
243 1. Fixed a potential buffer overflow in restore. Thanks
244 to Stan Bubrouski <satan@fastdial.net> for reporting
245 the bug.
246
247 2. Fixed a readline-related bug which prevented
248 'cat DUMPFILE | restore -i -f -' from working. Thanks
249 to Charles Karney <karney@users.sourceforge.net>
250 for the bug report.
251
252 3. Changed a few "panic" into "exit", causing restore to
253 be more stable against some attacks (like the last one
254 reported on Bugtraq, although the last version of restore
255 was not vulnerable - just dumped core). Thanks to
256 Andreas Hasenack <andreas@conectiva.com.br> for reporting
257 the bugs to me.
258
259 4. Removed the suid-root bit on dump and restore in the
260 default build (and generated RPMs). It should be safer
261 now. Users who need the suid-root capabilities in order
262 to make network backups should read first the man page
263 of dump and enable the suid bit manually.
264
265 5. Added -ltermcap to the compile parameters for restore
266 when using readline support, in order to make the compile
267 process work on some systems (SuSE ?). Thanks to
268 Patrik Schindler <poc@pocnet.net> for reporting the bug.
269
270 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
271 ====================================================================
272
273 1. The -F script is called now *only* at the end of a tape,
274 not at the end of the dump. Thanks to Christian Haul
275 <haul@informatik.tu-darmstadt.de> for the bug report.
276
277 Normally, the device name and the current volume number
278 are passed on the command line of the script. However,
279 if you want the old style script (with no arguments
280 passed on the command line), you can enable it in
281 configure with the --enable-oldstylefscript.
282
283 2. Use posix signal handling to preserve dump functionality
284 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
285 the patch.
286
287 3. Made the exit code of restore in case of a 'C'ompare
288 command reflect the result of the compare. An exit status
289 of 0 means the dump archive is correct, 1 means tape errors,
290 2 means that some files were modified. Thanks to Kenneth Porter
291 <shiva@well.com> for the suggestion.
292
293 4. Made (finally) quotes work in interactive restore.
294
295 5. Small fixes in order to allow dump to compile with a
296 really old e2fsprogs version. Thanks to Ian Zimmerman
297 <itz@speakeasy.org> for the bug report.
298
299 6. Add GNU readline capabilities to interactive restore.
300 Use configure's --enable-readline flag to enable this feature.
301 Thanks to Patrik Schindler <poc@pocnet.net> for the
302 suggestion.
303
304 7. Do the compare on the fly in restore 'C' mode (this will
305 allow not to exhaust the available /tmp space when
306 ccmparing large files). Thanks to Kenneth Porter
307 <shiva@well.com> for the suggestion.
308
309 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
310 ====================================================================
311
312 1. Fixed some several duplicate 'DUMP: DUMP:' in the
313 output of dump.
314
315 2. Corrected the estimation of blocks to dump. Note that
316 this is still wrong for large directory inodes, due
317 to the size of a BSD directory entry on the tape
318 (estimation is based on the size of the ext2 directory,
319 which is smaller because it doesn't contain the null
320 character at the end).
321
322 3. Made dump report the total number of MB written to
323 tape at the end of run. Thanks to W. Reilly Cooley
324 <wcooley@nakedape.cc> for the patch.
325
326 4. Added the -X option to restore giving the possibility
327 to read the names of the files to be extracted/listed
328 from a text file (in addition of the command line).
329 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
330 patch.
331
332 5. Added the device name and the current volume number
333 as arguments to the end of tape script (dump -F option).
334
335 6. Made the multi-volume dump work again (after having
336 broken it in 0.4b15).
337
338 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
339 ===================================================================
340
341 1. Added a prompt command in interactive restore mode. Thanks
342 to Andreas Dilger <adilger@home.com> for the patch.
343
344 2. Fixed a buffer overflow problem in dump (caused by
345 not checking the size of the filesystem parameter).
346 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
347 reporting this on Bugtraq (and to several dump users
348 who forwarded me his mail).
349
350 3. Added the '-F script' option to dump in order to
351 launch a script at the end of each tape (to be used
352 with a tape changer, or to notify the sysadmin by
353 pager etc.).
354
355 4. Fixed a bug in restore compare code caused by the changes
356 I made in 0.4b14.
357
358 5. Fixed the treatment of options using the old BSD syntax
359 in both dump and restore.
360
361 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
362 =======================================================================
363
364 1. Fixed a bug in dump which may have caused invalid deleted
365 directories to be dumped out if they were deleted after the
366 mapping pass. This could occure on active filesystem and lead
367 to heap corruption (causing dump malfunction in many possible ways).
368 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
369 bug and submitting the patch.
370
371 2. Fixed the handling of the filesystem parameter in dump. A
372 '/mnt/disk' parameter caused the disk contents to be dumped,
373 but a '/mnt/disk/' parameter caused the mountpoint directory
374 to be dumped (generally an empty directory).
375
376 3. Improved the output of dump in order to tell which directory
377 it is currently dumping (when dumping a subtree).
378
379 4. Added the '-e' exclude inode option to dump. Thanks to
380 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
381
382 5. Added a REPORTING-BUGS file in order to provide a guide
383 on how to correctly report a bug in dump/restore.
384
385 6. Fixed a restore problem when restoring a hard link to an inode
386 having the immutable or append-only attribute set. Thanks to
387 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
388
389 7. Fixed a compatibility problem between dumps made on little
390 endian machines (the format was correct) and big endian
391 machines (the format was incorrect). This fix break the
392 compatibility with the older dumps made on big endian
393 machines (sparc, m86k, ppc etc). For the first time in
394 linux dump's history (I believe), the dumps made by *BSD,
395 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
396 so interchangeable. Thanks to Rob Cermak
397 <cermak@ahab.rutgers.edu> for submitting the bug and
398 helping me test the fix.
399
400 8. Fixed the way dump reports the remaining percent/time, if
401 the number of blocks actually dumped exceeds the estimated
402 number of blocks. Thanks to Jean-Paul van der Jagt
403 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
404
405 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
406 ======================================================================
407
408 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
409 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
410
411 2. Removed the requirement to build the RPM as root from the
412 spec file. Thanks to Christian Weisgerber
413 <naddy@mips.rhein-neckar.de> for submitting this.
414
415 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
416 causing dump to block when dumping really huge filesystems.
417 Many thanks to Patrik Schindler <poc@pocnet.net> for
418 helping me find this bug.
419
420 4. Fixed the treatment for an interrupt signal when dump access
421 the remote tape through RSH. Thanks to Christian Weisgerber
422 <naddy@mips.rhein-neckar.de> for providing the patch.
423
424 5. Fixed a bug which was causing dump/restore to display
425 garbage characters instead of the remote host name.
426
427 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
428 =====================================================================
429
430 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
431 <kukuk@suse.de> for submitting the patch.
432
433 2. Fix for the exit code when using the size estimate option of
434 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
435 submitting the patch.
436
437 3. Handle EINTR in atomical reads/writes in dump, which was causing
438 dump to fail on some systems. Thanks to Eric Jergensen
439 <eric@dvns.com> for reporting the bug and submitting the patch.
440
441 4. Handle more than 16 characters for the device names in dumpdates.
442 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
443 tracking down the problem and proposing the solution.
444
445 5. Fixed a bug in dump which prevented the creation of the
446 dumpdates file when doing a 0-level dump without already
447 having a dumpdates file. Thanks to Patrik Schindler
448 <poc@pocnet.net> for reporting the bug.
449
450 6. Changed the way dump 'S' flag reports the size estimate
451 from number of blocks into bytes (making it compatible
452 with the Solaris version, and simplifying things for
453 amanda users). Thanks to Jason L Tibbitts III
454 <tibbs@math.uh.edu> for reporting the bug.
455
456 7. Fixed a compatibility problem in linux/alpha dump tape format.
457 Now the linux/alpha dump are (again) compatible with the
458 other dump formats. But this breaks compatibility with
459 older dumps made on alpha. Thanks to Mike Tibor
460 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
461
462 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
463 ======================================================================
464
465 1. Added a '--enable-kerberos' to configure.
466
467 2. Added a 'S' option to dump which determines the amount of space
468 that is needed to perform the dump without actually doing it, similar
469 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
470 <cermak@ahab.rutgers.edu>.
471
472 3. Added a 'M' multi-volume option to dump and restore which enables
473 dumping to multiple files (useful when dumping to an ext2
474 partition to make several dump files in order to bypass the 2GB
475 file size limitation). The 'f' argument is treated as a prefix and
476 the output files will be named <prefix>001, <prefix>002 etc. With
477 the 'M' flag, restore automatically selects the right file without
478 asking to enter a new tape each time.
479
480 4. Fixed a memory leak which was causing dump to grow very big
481 (270MB when dumping a 10GB filesystem...). Thanks to Jason
482 Fearon <jasonf@netrider.org.au> for reporting the bug.
483
484 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
485 ======================================================================
486
487 1. Make configure test if the system glob routines support
488 extended syntax (ALTDIRFUNC). If not, use the internal glob
489 routines instead of system ones. Thanks to Bernhard Sadlowski
490 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
491 and helping me resolve this and other minor libc5 compiling
492 glitches.
493
494 2. Fix a problem when dumping a ext2fs with the 'filetype'
495 feature enabled. Thanks to Patrick J. LoPresti
496 <patl@cag.lcs.mit.edu> for reporting the bug and to
497 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
498
499 3. Made the nodump flag work on directories. A directory which
500 has the nodump flag gets never dumped, regardless of its
501 contents.
502
503 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
504 which allows dump on an active ext3 filesystem. However, this
505 is a "quick and dirty" patch which enables backup of an ext3
506 filesystem through the ext2 compatibility (by ignoring the
507 NEEDS_RECOVERY bit). The journal file is not recognized and
508 it is dumped (it should not).
509
510 5. Test the superblock compatibility flags when dumping, in order
511 to be sure that we know how to deal with specific features.
512
513 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
514 ====================================================================
515
516 1. Use lchown instead of chown, fixing a possible security problem
517 when restoring symlinks (a malicious user could use this
518 to deliberately corrupt the ownership of important system files).
519 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
520 this and providing the patch.
521
522 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
523 ====================================================================
524
525 1. Put dump sources under CVS, added Id tags in all files so
526 one can use 'ident' on binary files.
527
528 2. Added the dump/restore version in the usage text so one can
529 easily verify the version he is using.
530
531 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
532 a va_start/va_end problem on linux-ppc (always call va_start
533 va_end in pairs each time we use a vararg function).
534
535 4. Added again the DT_* constants because old libc does not
536 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
537 for submitting the bug report.
538
539 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
540 this should enable dumping big (huge) filesystems.
541
542 6. Added the RSH environment variable in order to be able to
543 use a rsh replacement like ssh when doing remote backups (and
544 bypass the security limitations of rcmd). Now you can do remote
545 backups without being root (or making dump setuid root).
546
547 7. Modified again the way dumpdates works. For incremental dumps,
548 we need to read dumpdates even if we are not using 'u' option.
549 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
550 this should work.
551
552 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
553 ===================================================================
554
555 1. Removed the 'k' flag from the restore 'about' text if kerberos
556 was not compiled in.
557
558 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
559 to them (fsetflags takes a char*, setflags an open fd!).
560
561 3. (f)setflags is called only if the flags aren't empty. If the
562 file is a special file, a warning is printed, because changing
563 flags implies opening the device. Normally, a special file
564 should not have any flag... (Debian bug #29775, patch provided
565 by Abhijit Dasgupta <abhijit@ans.net>).
566
567 4. Made possible to dump a file system not mentioned in /etc/fstab.
568 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
569
570 5. Changed the default behaviour to not create dumpdates
571 unless 'u' option is specified. Removed the old "debian-patch"
572 which provided the same thing. (Debian bug #38136, #33818).
573
574 6. Removed all those dump*announce, since they were getting old...
575
576 7. Added warning messages if dumpdates does not exist and
577 when an inferior level dump does not exist (except for a level 0
578 dump).
579
580 8. Debugged the glob calls in interactive mode: restore used a
581 dirent struct which was different from the /usr/include/dirent.h
582 one (this used to work, is it a glibc2 change?), so none of the
583 compat glob (which used /usr/include/dirent.h) or the system glob
584 worked. Restore use now the system dirent (and the system
585 DT_* constants), which are compatible with BSD ones.
586
587 9. Added a configure flag (--with-dumpdatespath) to specify
588 the location of dumpdates file. By default, it is
589 /etc/dumpdates.
590
591 10. Added the "AUTHOR" and "AVAILABILITY" sections and
592 included the current date/version in man pages.
593
594 11. Corrected the estimation of remaining time when
595 the operator doesn't change the tapes quickly enough. This
596 was an old bug, I thought I corrected it, and discovered
597 that in fact it was corrected in two different places, so
598 the results canceled each other...
599
600 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
601 ===================================================================
602
603 1. Integrated multiple patches from RedHat, Debian and SuSE:
604
605 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
606 isn't specified.
607 - several fixes in the man pages.
608 - update the default tape device to /dev/st0.
609 - many updates for Linux Alpha (byte ordering, size_t etc).
610 - buffer overruns.
611 - use environment variable for TMPDIR (instead of /tmp).
612 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
613 - workaround egcs bug (RedHat bugs #4281 and #2989).
614 - wire $(OPT) throughout Makefile's.
615
616 2. Upgrade the dump revision to 1, making possible to dump filesystems
617 made with e2fsprogs-1.15 or newer. Nothing seems to break...
618
619 3. Fix some compile warnings, prototype all functions.
620
621 4. Use glibc err/glob instead of internal compatibility
622 routines (only if available).
623
624 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
625 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
626
627 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
628 ======================================================================
629
630 1. Integrated the changes from FreeBSD-3.1-RELEASE
631 (mostly bug fixes, buffer overruns, dump has now an "automatic
632 tape length calculation" flag, dump/restore can use kerberos now
633 (this is NOT tested), use environment variables for TAPE and
634 RMT etc.).
635
636 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
637
638 3. Corrected a bug in restore when using 'C' option with multi-volumes
639 tapes (files splited accros two tapes give "size changed" errors
640 when comparing).
641
642 4. Corrected the long standing bug when dumping multiple tapes.
643 This works for me, needs further testing.
644
645 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
646 ====================================================================
647
648 1. Dump now runs correctly on kernels 2.1.x
649 Fix made by Gerald Peters <gapeters@worldnet.att.net>
650
651 Changes between versions 0.4b2 and 0.4b3
652 ========================================
653
654 1. Use realpath() if available
655
656 2. Report statistics
657
658 Changes between versions 0.4b1 and 0.4b2
659 ========================================
660
661 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
662 the patch)
663
664 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
665
666 3. Fixed configure.in to correctly handle --enable-rmt
667
668 Changes between versions 0.3 and 0.4b1
669 ======================================
670
671 1. Integrated the changes from 4.4BSD-Lite2
672
673 2. Integrated the patches from Debian and Red Hat
674
675 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
676
677 4. Changed dump to use the Ext2fs library to get block addresses. This
678 should solve the endianness problem on SparcLinux.
679
680 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
681 distribution's one) to use autoconf
682
683 6. Fixed a few minor bugs
684
685 Changes between versions 0.2e and 0.2f
686 ======================================
687
688 1. Added the creation of named pipes (fifos) in restore.
689
690 2. Added the -N flag in the restore manual page.
691
692 3. Added the file kernel.patch which contains the llseek() optimization
693 patch for 1.2.x kernels.
694
695 4. Fixed a bug in the restoration of symbolic links: owner and group were
696 not restored.
697
698 5. Integrated some changes from FreeBSD 2.2.
699
700 6. Added a call to ftruncate() after restoring each file to restore
701 correctly files ending by a hole.
702
703 Changes between versions 0.2d and 0.2e
704 ======================================
705
706 1. Fixed a bug in the "set owner/mode" process. Every file was restored
707 with owner = root (0) and group = root/wheel/whatever (0).
708
709 Changes between versions 0.2c and 0.2d
710 ======================================
711
712 1. Dump is now able to backup 2GB+ filesystems.
713
714 2. Dump and restore can now be linked as static binaries.
715
716 Changes between versions 0.2b and 0.2c
717 ======================================
718
719 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
720
721 Changes between versions 0.2a and 0.2b
722 ======================================
723
724 1. Really fixed the bug that I should have corrected in 0.2a.
725
726 2. Enabled optimization again.
727
728 Changes between versions 0.2 and 0.2a
729 =====================================
730
731 1. Disabled the optimization during compilation.
732
733 Changes between versions 0.1 and 0.2
734 ====================================
735
736 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
737 the fs_type field (actually, I modified the file fstab.c to make it
738 use the mntent functions).
739
740 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
741 made by dump should be readable by the BSD restore and backups made
742 by the BSD dump should be readable by restore. Unfortunately, this
743 means that the dump format has changed between version 0.1 and version
744 0.2 :-(
745
746 3. Dump is now able to backup a subtree, it is no longer limited to whole
747 filesystems like the BSD version.
748
749 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
750 than 2 GB.
751
752 Changes between versions 0.0 and 0.1
753 ====================================
754
755 1. Now create links rdump and rrestore during the `make install' step.
756
757 2. Linux port specific bugs added to the manual pages
758
759 3. Incorrect estimation of the number of tapes blocks fixed when doing
760 incremental backups.
761
762 4. Better ls-like format in restore in interactive mode.