]> git.wh0rd.org - dump.git/blob - CHANGES
a debug message removed
[dump.git] / CHANGES
1 $Id: CHANGES,v 1.142 2001/11/13 12:11:05 stelian Exp $
2
3 Changes between versions 0.4b24 and 0.4b25 (released ??????????????????)
4 ========================================================================
5
6 1. Added a mini howto from Patrick Walsh in the examples
7 directory.
8
9 2. Minor man pages syntax corrections. Thanks to
10 Chris Danis <screechco@home.com> for reporting the bugs.
11
12 3. Added a script from David B. Peterson <dave@toppledwagon.com>
13 to the examples section. It features dumping several
14 filesystems to a remote tape drive upon ssh.
15
16 4. Added a patch provided by Richard Jones <rich@annexia.org>
17 which allows BRADEMAX (number of read errors tolerated by
18 dump) to be adjusted using the -I option.
19
20 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
21 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
22 patch.
23
24 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
25 ========================================================================
26
27 1. Fixed the permissions of a newly created QFA file by dump.
28
29 2. Cleaned up the source of dump (the external variables
30 definition was a complete mess, making possible to have
31 objects overlap).
32
33 3. Fixed restore to use the full tape volume path when doing
34 a compare (since it changes the working directory to the
35 filesystem being compared in the process).
36
37 4. Added the -q option to dump which makes dump abort
38 whenever it needs operator attention. It should be
39 easier to use dump in scripts with this option.
40
41 5. Detect the use of incompatible options to dump and
42 refuse them (like -a and -B options together).
43
44 6. Added bzip2 compression to dump/restore (use option -j level
45 to select it). Note that you will need a restore version
46 >= 0.4b24 in order to restore a bzip2 compressed dump.
47 The same warning as for the zlib compression applies:
48 the tape format of a bzip2 dump is not compatible with the
49 original BSD tape format.
50
51 7. Fixed a overflow problem in dump corrupting the dump when
52 very large files were encountered. Thanks to Vladimir Nadvornik
53 for the bug report.
54
55 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
56 kernel buffer/page cache before starting the dump, helping
57 a bit those who use dump on mounted filesystems. Thanks to
58 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
59 for this suggestion.
60
61 9. Updated the RPM spec file following the RedHat changes
62 (dynamically linked binaries now in /usr/sbin etc).
63
64 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
65 which enables restore to recognise multi volume compressed dumps
66 done on CD. Included his scripts for dump (which pipe the dump
67 output directly into cdrecord) and restore. There is now
68 possible to to backups to CD on the fly!
69
70 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
71 ===================================================================
72
73 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
74 Marcus Meissner (Caldera International Security Dept.).
75
76 2. Implement the Sun rmt extended protocol. Patch submitted
77 by Ian Gordon <iangordon@users.sourceforge.net>.
78
79 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
80 the linux kernel header. This ensures that dump always has the
81 latest version of this file. Patch submitted by
82 Andreas Dilger <adilger@turbolinux.com>.
83
84 4. Report any filesystem present in either /etc/fstab with a
85 positive passno or /etc/dumpdates in dump -w output.
86 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
87
88 5. Fixed the looping problem in dump introduced in the
89 previous version.
90
91 6. Changed the -B option of dump to limit the size of
92 _compressed_ output per volume if compression is on.
93 Patch contributed by Helmut Jarausch
94 <jarausch@igpm.rwth-aachen.de>. Note however that, since
95 it is impossible to predict the size of the compressed
96 data before doing the compression, the -B limit is a bit
97 conservative.
98
99 7. Fixed a bug in reading the operator typed file/tape path for
100 the next volume in restore.
101
102 8. Implemented a "-F script" option to restore which permits the
103 user to specify a script which will be launched at the
104 beginning of each tape, useful for automatic programming of
105 tape changers for example. See the restore man page for the
106 script parameters and return codes.
107
108 9. Small fix for the QFA routines provided by Uwe Gohlke
109 <uwe@ugsoft.de>, and some recommendations for QFA uses in
110 the man pages.
111
112 10. Fixed the multivolume restoring where making a mistake
113 to the 'Mount next tape' prompt caused several blocks to
114 be lost.
115
116 11. Enhanced the -e option of dump to take as a parameter a
117 comma separated list of inode numbers.
118
119 12. Added the -E option to dump which specify a file containing
120 inode numbers to exclude from the dump.
121
122 13. Fixed the compressed multi-volume dump + restore.
123
124 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
125 ==================================================================
126
127 1. Made dump report the number of blocks dumped per volume.
128 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
129
130 2. Fix a bug in dump where waiting too much at the 'change volumes'
131 question caused the volume to be rewritten. Thanks to
132 Graham Reed <greed@users.sourceforge.net> for reporting the
133 bug and providing a patch.
134
135 3. Added a compression option to dump, contributed by Jerry
136 Peters <gapeters@worldnet.att.net>.
137
138 WARNING: the compression works only on files/pipes and
139 drives supporting variable block size.
140
141 WARNING2: tapes created using the compression feature are
142 incompatible with the standard dump tape format, and a
143 version of restore >= 0.4b22 will be needed for extraction.
144
145 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
146 architectures. Thanks to Paul Slootman <paul@debian.org> for
147 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
148 it upstream.
149
150 5. Many cleanups (CPP defines, const char warnings, check of
151 ext2fs COMPAT flags, time_t cleanups, added libext2 version
152 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
153
154 6. Made --prefix option work in configure. All the install path
155 are now based on the configure parameters.
156
157 7. Added the Quick File Access mode in dump/restore, contributed
158 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
159 a file tape position for each inode, and this file is used by
160 restore (if called with parameter Q and the filename)
161 to directly position the tape at the file restore is currently
162 working on. This saves hours when restoring single files from
163 large backups, saves the tapes and the drive's head. Use
164 --enable-qfa option of configure to compile in the QFA support.
165
166 8. Added the possibility to dump several files and directories
167 in a single invocation of dump. Thanks to Uwe Gohlke
168 <uwe@ugsoft.de> for implementing this option.
169
170 9. Fixed the dumping and restoring of files with holes
171 (files ending with holes didn't get dumped or restored
172 correctly).
173
174 10. Fixed a socket descriptor leak in dump, which leaved opened
175 3 file descriptors per dump process (and there is one dump
176 process per tape).
177
178 11. Fixed dump large file system support, by explicit use of
179 open64/lseek64/etc functions (explicit use needed because
180 e2fsprogs libraries don't behave well when compiled with
181 FILE_OFFSET_BITS=64).
182
183 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
184 ======================================================================
185
186 1. Fixed some bugs in the dump -w|-W logic introduced by
187 the previous version. Thanks to Andreas Dilger
188 <adilger@turbolinux.com> for his help on this one.
189
190 2. Fixed again a compilation problem when using old e2fs
191 headers (filesystem label related). Thanks to many users
192 who reported this stupid error.
193
194 3. Fixed a build problem on old lib5 based systems dealing with
195 _PATH_MNTTAB being undefined. Thanks to John Adams
196 <johna@onevista.com> for reporting the bug.
197
198 4. Improved the error detection in restore when creating
199 the temporary files in TMPDIR. Now restore will corectly
200 report a 'no space left on device' error instead of
201 strange errors which could imply an archive corruption.
202 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
203 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
204
205 5. Added the throughput information in dump's progression
206 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
207 for the patch.
208
209 6. Use libext2fs's inode scanning routines, which are particularly
210 robust when dealing with errors (especially when having some
211 bad blocks on the filesystem). Many thanks to Colin
212 <colin@horizon.com> for reporting the bug, investigating
213 the issues, finding a workaround, writing the patch and
214 fully testing it... (of course, if this patch will break
215 anything, he is to blame for :-)).
216
217 7. Made dump and restore LFS aware. Dump can dump a filesystem
218 containing large files, generate a large file on output and
219 restore can restore them. This feature is not enabled by
220 default, use --enable-largefile option of configure to enable
221 it (you need to have a LFS aware glibc though). Thanks to
222 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
223 and to Theodore T'so <tytso@valinux.com> for his always
224 useful thoughts.
225
226 8. Made dump ask upon a tape write error if it should rewrite
227 the current volume (assume this is a bad tape) or if it should
228 assume an end-of-tape condition (useful for tape drives which
229 behaves badly at the end of the tape). Thanks to Andreas
230 Dilger <adilger@turbolinux.com> for the suggestion.
231
232 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
233 =======================================================================
234
235 1. Fixed a small compilation problem due to a change
236 in the definintion of the struct sigaction in
237 glibc 2.0 / libc5. Thanks to Gunther Schlegel
238 <schlegel@riege.de> for reporting the bug and to
239 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
240 a fix.
241
242 2. Modified the label and uuid analysis in order to be
243 self-contained (independant of kernel/libc headers). This
244 should fix the compile with older kernel/libc headers and
245 will preserve the functionality. Thanks to Bernhard Erdmann
246 <bernhard.erdmann@gmx.de> for reporting the bug.
247
248 3. The 'exclude inode' option, if applied to a directory
249 inode, excludes now correctly all the leaves of this
250 directory. Thanks to John R. Dennison
251 <gerdesas@users.sourceforge.net> for reporting the bug.
252
253 4. Fixed the '-e' option to disable the possibility
254 to exclude the root inode (which causes the dumps to
255 be unreadable by restore). Prevented array overflow
256 when multiple -e options are used.
257
258 5. Fixed dump to correctly interpret a filesystem argument
259 which is a mountpoint and it is not an absolute path
260 (as specified in the fstab). Thanks to Bernhard R. Erdmann
261 <be@berdmann.de> for reporting the bug.
262
263 6. Made dump able to backup files larger than 2 GB. Note that
264 dump still doesn't cope with files larger than 4 GB.
265
266 7. Restore the real uid and gid before invoking an external
267 RSH program (big hole when dump or restore is suid root!).
268
269 8. Get the values for the default tape device and the location
270 of fstab file from the system headers. Thanks to
271 Andreas Dilger <adilger@turbolinux.com> for the patch.
272
273 9. Made dump -w|-W report all recognized filesystems
274 present in either /etc/fstab or /etc/dumpdates, and present
275 the list in the same order as in fstab file. Thanks
276 to Andreas Dilger <adilger@turbolinux.com> for the patch.
277
278 10. Made dump's -a (automatic end of tape detection) the
279 default. Specifying one of -B, -c, -d or -s options will
280 override the EOM detection. Thanks to Andreas Dilger
281 <adilger@turbolinux.com> for the patch.
282
283 11. Save the ext2 filesystem label into the dump volume label.
284 Specifying a volume label on the command line (-L option)
285 will override this feature. Thanks to Andreas Dilger
286 <adilger@turbolinux.com> for the patch.
287
288 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
289 =====================================================================
290
291 1. Fixed the signal handling in dump (which I broke in 0.4b17)
292 which was causing several strange dump failures (dump
293 hanged or segmentation faults etc).
294
295 2. Specified the default blocksize in the dump man page.
296
297 3. Changed two info messages of restore to be written on stdout
298 instead of stderr in order to leave stderr only for errors
299 or warnings. Thanks to Stephen Atwell
300 <satwell@urbana.css.mot.com> for the suggestion.
301
302 4. Corrected an off by one calculation which prevented
303 dumping a directory having a 1 character name length.
304 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
305 for reporting the bug.
306
307 5. Reinforce dump in order to allow only level 0 and no
308 -u option when dumping a subdirectory, and document
309 this in the man page. Thanks to Bernhard Erdmann
310 <bernhard.erdmann@gmx.de> for reporting the bug.
311
312 6. Fixed a small harmless bug in restore which caused
313 in some conditions hard links to be restored several
314 times (therefore generation some warning messages).
315 Thanks to Kenneth Porter <shiva@well.com> for
316 reporting the bug.
317
318 7. Updated the RPM spec file to the latest RedHat version,
319 providing FHS packaging and other cosmetic changes.
320 You will need RPM version >= 3.0.5 in order to build the RPM.
321
322 8. Updated the configure script to check for libtermcap
323 before checking for libreadline, because we need this
324 library in order to compile the readline support.
325
326 9. Made dump understand the LABEL= and UUID= notation
327 both in /etc/fstab and on the command line. Note that
328 you will need the /proc filesystem in order to use
329 these notations. Thanks to Erik Troan <ewt@redhat.com>
330 for providing the patch.
331
332 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
333 ===================================================================
334
335 1. Fixed a potential buffer overflow in restore. Thanks
336 to Stan Bubrouski <satan@fastdial.net> for reporting
337 the bug.
338
339 2. Fixed a readline-related bug which prevented
340 'cat DUMPFILE | restore -i -f -' from working. Thanks
341 to Charles Karney <karney@users.sourceforge.net>
342 for the bug report.
343
344 3. Changed a few "panic" into "exit", causing restore to
345 be more stable against some attacks (like the last one
346 reported on Bugtraq, although the last version of restore
347 was not vulnerable - just dumped core). Thanks to
348 Andreas Hasenack <andreas@conectiva.com.br> for reporting
349 the bugs to me.
350
351 4. Removed the suid-root bit on dump and restore in the
352 default build (and generated RPMs). It should be safer
353 now. Users who need the suid-root capabilities in order
354 to make network backups should read first the man page
355 of dump and enable the suid bit manually.
356
357 5. Added -ltermcap to the compile parameters for restore
358 when using readline support, in order to make the compile
359 process work on some systems (SuSE ?). Thanks to
360 Patrik Schindler <poc@pocnet.net> for reporting the bug.
361
362 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
363 ====================================================================
364
365 1. The -F script is called now *only* at the end of a tape,
366 not at the end of the dump. Thanks to Christian Haul
367 <haul@informatik.tu-darmstadt.de> for the bug report.
368
369 Normally, the device name and the current volume number
370 are passed on the command line of the script. However,
371 if you want the old style script (with no arguments
372 passed on the command line), you can enable it in
373 configure with the --enable-oldstylefscript.
374
375 2. Use posix signal handling to preserve dump functionality
376 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
377 the patch.
378
379 3. Made the exit code of restore in case of a 'C'ompare
380 command reflect the result of the compare. An exit status
381 of 0 means the dump archive is correct, 1 means tape errors,
382 2 means that some files were modified. Thanks to Kenneth Porter
383 <shiva@well.com> for the suggestion.
384
385 4. Made (finally) quotes work in interactive restore.
386
387 5. Small fixes in order to allow dump to compile with a
388 really old e2fsprogs version. Thanks to Ian Zimmerman
389 <itz@speakeasy.org> for the bug report.
390
391 6. Add GNU readline capabilities to interactive restore.
392 Use configure's --enable-readline flag to enable this feature.
393 Thanks to Patrik Schindler <poc@pocnet.net> for the
394 suggestion.
395
396 7. Do the compare on the fly in restore 'C' mode (this will
397 allow not to exhaust the available /tmp space when
398 ccmparing large files). Thanks to Kenneth Porter
399 <shiva@well.com> for the suggestion.
400
401 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
402 ====================================================================
403
404 1. Fixed some several duplicate 'DUMP: DUMP:' in the
405 output of dump.
406
407 2. Corrected the estimation of blocks to dump. Note that
408 this is still wrong for large directory inodes, due
409 to the size of a BSD directory entry on the tape
410 (estimation is based on the size of the ext2 directory,
411 which is smaller because it doesn't contain the null
412 character at the end).
413
414 3. Made dump report the total number of MB written to
415 tape at the end of run. Thanks to W. Reilly Cooley
416 <wcooley@nakedape.cc> for the patch.
417
418 4. Added the -X option to restore giving the possibility
419 to read the names of the files to be extracted/listed
420 from a text file (in addition of the command line).
421 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
422 patch.
423
424 5. Added the device name and the current volume number
425 as arguments to the end of tape script (dump -F option).
426
427 6. Made the multi-volume dump work again (after having
428 broken it in 0.4b15).
429
430 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
431 ===================================================================
432
433 1. Added a prompt command in interactive restore mode. Thanks
434 to Andreas Dilger <adilger@home.com> for the patch.
435
436 2. Fixed a buffer overflow problem in dump (caused by
437 not checking the size of the filesystem parameter).
438 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
439 reporting this on Bugtraq (and to several dump users
440 who forwarded me his mail).
441
442 3. Added the '-F script' option to dump in order to
443 launch a script at the end of each tape (to be used
444 with a tape changer, or to notify the sysadmin by
445 pager etc.).
446
447 4. Fixed a bug in restore compare code caused by the changes
448 I made in 0.4b14.
449
450 5. Fixed the treatment of options using the old BSD syntax
451 in both dump and restore.
452
453 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
454 =======================================================================
455
456 1. Fixed a bug in dump which may have caused invalid deleted
457 directories to be dumped out if they were deleted after the
458 mapping pass. This could occure on active filesystem and lead
459 to heap corruption (causing dump malfunction in many possible ways).
460 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
461 bug and submitting the patch.
462
463 2. Fixed the handling of the filesystem parameter in dump. A
464 '/mnt/disk' parameter caused the disk contents to be dumped,
465 but a '/mnt/disk/' parameter caused the mountpoint directory
466 to be dumped (generally an empty directory).
467
468 3. Improved the output of dump in order to tell which directory
469 it is currently dumping (when dumping a subtree).
470
471 4. Added the '-e' exclude inode option to dump. Thanks to
472 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
473
474 5. Added a REPORTING-BUGS file in order to provide a guide
475 on how to correctly report a bug in dump/restore.
476
477 6. Fixed a restore problem when restoring a hard link to an inode
478 having the immutable or append-only attribute set. Thanks to
479 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
480
481 7. Fixed a compatibility problem between dumps made on little
482 endian machines (the format was correct) and big endian
483 machines (the format was incorrect). This fix break the
484 compatibility with the older dumps made on big endian
485 machines (sparc, m86k, ppc etc). For the first time in
486 linux dump's history (I believe), the dumps made by *BSD,
487 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
488 so interchangeable. Thanks to Rob Cermak
489 <cermak@ahab.rutgers.edu> for submitting the bug and
490 helping me test the fix.
491
492 8. Fixed the way dump reports the remaining percent/time, if
493 the number of blocks actually dumped exceeds the estimated
494 number of blocks. Thanks to Jean-Paul van der Jagt
495 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
496
497 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
498 ======================================================================
499
500 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
501 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
502
503 2. Removed the requirement to build the RPM as root from the
504 spec file. Thanks to Christian Weisgerber
505 <naddy@mips.rhein-neckar.de> for submitting this.
506
507 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
508 causing dump to block when dumping really huge filesystems.
509 Many thanks to Patrik Schindler <poc@pocnet.net> for
510 helping me find this bug.
511
512 4. Fixed the treatment for an interrupt signal when dump access
513 the remote tape through RSH. Thanks to Christian Weisgerber
514 <naddy@mips.rhein-neckar.de> for providing the patch.
515
516 5. Fixed a bug which was causing dump/restore to display
517 garbage characters instead of the remote host name.
518
519 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
520 =====================================================================
521
522 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
523 <kukuk@suse.de> for submitting the patch.
524
525 2. Fix for the exit code when using the size estimate option of
526 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
527 submitting the patch.
528
529 3. Handle EINTR in atomical reads/writes in dump, which was causing
530 dump to fail on some systems. Thanks to Eric Jergensen
531 <eric@dvns.com> for reporting the bug and submitting the patch.
532
533 4. Handle more than 16 characters for the device names in dumpdates.
534 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
535 tracking down the problem and proposing the solution.
536
537 5. Fixed a bug in dump which prevented the creation of the
538 dumpdates file when doing a 0-level dump without already
539 having a dumpdates file. Thanks to Patrik Schindler
540 <poc@pocnet.net> for reporting the bug.
541
542 6. Changed the way dump 'S' flag reports the size estimate
543 from number of blocks into bytes (making it compatible
544 with the Solaris version, and simplifying things for
545 amanda users). Thanks to Jason L Tibbitts III
546 <tibbs@math.uh.edu> for reporting the bug.
547
548 7. Fixed a compatibility problem in linux/alpha dump tape format.
549 Now the linux/alpha dump are (again) compatible with the
550 other dump formats. But this breaks compatibility with
551 older dumps made on alpha. Thanks to Mike Tibor
552 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
553
554 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
555 ======================================================================
556
557 1. Added a '--enable-kerberos' to configure.
558
559 2. Added a 'S' option to dump which determines the amount of space
560 that is needed to perform the dump without actually doing it, similar
561 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
562 <cermak@ahab.rutgers.edu>.
563
564 3. Added a 'M' multi-volume option to dump and restore which enables
565 dumping to multiple files (useful when dumping to an ext2
566 partition to make several dump files in order to bypass the 2GB
567 file size limitation). The 'f' argument is treated as a prefix and
568 the output files will be named <prefix>001, <prefix>002 etc. With
569 the 'M' flag, restore automatically selects the right file without
570 asking to enter a new tape each time.
571
572 4. Fixed a memory leak which was causing dump to grow very big
573 (270MB when dumping a 10GB filesystem...). Thanks to Jason
574 Fearon <jasonf@netrider.org.au> for reporting the bug.
575
576 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
577 ======================================================================
578
579 1. Make configure test if the system glob routines support
580 extended syntax (ALTDIRFUNC). If not, use the internal glob
581 routines instead of system ones. Thanks to Bernhard Sadlowski
582 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
583 and helping me resolve this and other minor libc5 compiling
584 glitches.
585
586 2. Fix a problem when dumping a ext2fs with the 'filetype'
587 feature enabled. Thanks to Patrick J. LoPresti
588 <patl@cag.lcs.mit.edu> for reporting the bug and to
589 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
590
591 3. Made the nodump flag work on directories. A directory which
592 has the nodump flag gets never dumped, regardless of its
593 contents.
594
595 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
596 which allows dump on an active ext3 filesystem. However, this
597 is a "quick and dirty" patch which enables backup of an ext3
598 filesystem through the ext2 compatibility (by ignoring the
599 NEEDS_RECOVERY bit). The journal file is not recognized and
600 it is dumped (it should not).
601
602 5. Test the superblock compatibility flags when dumping, in order
603 to be sure that we know how to deal with specific features.
604
605 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
606 ====================================================================
607
608 1. Use lchown instead of chown, fixing a possible security problem
609 when restoring symlinks (a malicious user could use this
610 to deliberately corrupt the ownership of important system files).
611 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
612 this and providing the patch.
613
614 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
615 ====================================================================
616
617 1. Put dump sources under CVS, added Id tags in all files so
618 one can use 'ident' on binary files.
619
620 2. Added the dump/restore version in the usage text so one can
621 easily verify the version he is using.
622
623 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
624 a va_start/va_end problem on linux-ppc (always call va_start
625 va_end in pairs each time we use a vararg function).
626
627 4. Added again the DT_* constants because old libc does not
628 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
629 for submitting the bug report.
630
631 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
632 this should enable dumping big (huge) filesystems.
633
634 6. Added the RSH environment variable in order to be able to
635 use a rsh replacement like ssh when doing remote backups (and
636 bypass the security limitations of rcmd). Now you can do remote
637 backups without being root (or making dump setuid root).
638
639 7. Modified again the way dumpdates works. For incremental dumps,
640 we need to read dumpdates even if we are not using 'u' option.
641 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
642 this should work.
643
644 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
645 ===================================================================
646
647 1. Removed the 'k' flag from the restore 'about' text if kerberos
648 was not compiled in.
649
650 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
651 to them (fsetflags takes a char*, setflags an open fd!).
652
653 3. (f)setflags is called only if the flags aren't empty. If the
654 file is a special file, a warning is printed, because changing
655 flags implies opening the device. Normally, a special file
656 should not have any flag... (Debian bug #29775, patch provided
657 by Abhijit Dasgupta <abhijit@ans.net>).
658
659 4. Made possible to dump a file system not mentioned in /etc/fstab.
660 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
661
662 5. Changed the default behaviour to not create dumpdates
663 unless 'u' option is specified. Removed the old "debian-patch"
664 which provided the same thing. (Debian bug #38136, #33818).
665
666 6. Removed all those dump*announce, since they were getting old...
667
668 7. Added warning messages if dumpdates does not exist and
669 when an inferior level dump does not exist (except for a level 0
670 dump).
671
672 8. Debugged the glob calls in interactive mode: restore used a
673 dirent struct which was different from the /usr/include/dirent.h
674 one (this used to work, is it a glibc2 change?), so none of the
675 compat glob (which used /usr/include/dirent.h) or the system glob
676 worked. Restore use now the system dirent (and the system
677 DT_* constants), which are compatible with BSD ones.
678
679 9. Added a configure flag (--with-dumpdatespath) to specify
680 the location of dumpdates file. By default, it is
681 /etc/dumpdates.
682
683 10. Added the "AUTHOR" and "AVAILABILITY" sections and
684 included the current date/version in man pages.
685
686 11. Corrected the estimation of remaining time when
687 the operator doesn't change the tapes quickly enough. This
688 was an old bug, I thought I corrected it, and discovered
689 that in fact it was corrected in two different places, so
690 the results canceled each other...
691
692 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
693 ===================================================================
694
695 1. Integrated multiple patches from RedHat, Debian and SuSE:
696
697 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
698 isn't specified.
699 - several fixes in the man pages.
700 - update the default tape device to /dev/st0.
701 - many updates for Linux Alpha (byte ordering, size_t etc).
702 - buffer overruns.
703 - use environment variable for TMPDIR (instead of /tmp).
704 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
705 - workaround egcs bug (RedHat bugs #4281 and #2989).
706 - wire $(OPT) throughout Makefile's.
707
708 2. Upgrade the dump revision to 1, making possible to dump filesystems
709 made with e2fsprogs-1.15 or newer. Nothing seems to break...
710
711 3. Fix some compile warnings, prototype all functions.
712
713 4. Use glibc err/glob instead of internal compatibility
714 routines (only if available).
715
716 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
717 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
718
719 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
720 ======================================================================
721
722 1. Integrated the changes from FreeBSD-3.1-RELEASE
723 (mostly bug fixes, buffer overruns, dump has now an "automatic
724 tape length calculation" flag, dump/restore can use kerberos now
725 (this is NOT tested), use environment variables for TAPE and
726 RMT etc.).
727
728 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
729
730 3. Corrected a bug in restore when using 'C' option with multi-volumes
731 tapes (files splited accros two tapes give "size changed" errors
732 when comparing).
733
734 4. Corrected the long standing bug when dumping multiple tapes.
735 This works for me, needs further testing.
736
737 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
738 ====================================================================
739
740 1. Dump now runs correctly on kernels 2.1.x
741 Fix made by Gerald Peters <gapeters@worldnet.att.net>
742
743 Changes between versions 0.4b2 and 0.4b3
744 ========================================
745
746 1. Use realpath() if available
747
748 2. Report statistics
749
750 Changes between versions 0.4b1 and 0.4b2
751 ========================================
752
753 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
754 the patch)
755
756 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
757
758 3. Fixed configure.in to correctly handle --enable-rmt
759
760 Changes between versions 0.3 and 0.4b1
761 ======================================
762
763 1. Integrated the changes from 4.4BSD-Lite2
764
765 2. Integrated the patches from Debian and Red Hat
766
767 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
768
769 4. Changed dump to use the Ext2fs library to get block addresses. This
770 should solve the endianness problem on SparcLinux.
771
772 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
773 distribution's one) to use autoconf
774
775 6. Fixed a few minor bugs
776
777 Changes between versions 0.2e and 0.2f
778 ======================================
779
780 1. Added the creation of named pipes (fifos) in restore.
781
782 2. Added the -N flag in the restore manual page.
783
784 3. Added the file kernel.patch which contains the llseek() optimization
785 patch for 1.2.x kernels.
786
787 4. Fixed a bug in the restoration of symbolic links: owner and group were
788 not restored.
789
790 5. Integrated some changes from FreeBSD 2.2.
791
792 6. Added a call to ftruncate() after restoring each file to restore
793 correctly files ending by a hole.
794
795 Changes between versions 0.2d and 0.2e
796 ======================================
797
798 1. Fixed a bug in the "set owner/mode" process. Every file was restored
799 with owner = root (0) and group = root/wheel/whatever (0).
800
801 Changes between versions 0.2c and 0.2d
802 ======================================
803
804 1. Dump is now able to backup 2GB+ filesystems.
805
806 2. Dump and restore can now be linked as static binaries.
807
808 Changes between versions 0.2b and 0.2c
809 ======================================
810
811 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
812
813 Changes between versions 0.2a and 0.2b
814 ======================================
815
816 1. Really fixed the bug that I should have corrected in 0.2a.
817
818 2. Enabled optimization again.
819
820 Changes between versions 0.2 and 0.2a
821 =====================================
822
823 1. Disabled the optimization during compilation.
824
825 Changes between versions 0.1 and 0.2
826 ====================================
827
828 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
829 the fs_type field (actually, I modified the file fstab.c to make it
830 use the mntent functions).
831
832 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
833 made by dump should be readable by the BSD restore and backups made
834 by the BSD dump should be readable by restore. Unfortunately, this
835 means that the dump format has changed between version 0.1 and version
836 0.2 :-(
837
838 3. Dump is now able to backup a subtree, it is no longer limited to whole
839 filesystems like the BSD version.
840
841 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
842 than 2 GB.
843
844 Changes between versions 0.0 and 0.1
845 ====================================
846
847 1. Now create links rdump and rrestore during the `make install' step.
848
849 2. Linux port specific bugs added to the manual pages
850
851 3. Incorrect estimation of the number of tapes blocks fixed when doing
852 incremental backups.
853
854 4. Better ls-like format in restore in interactive mode.