]> git.wh0rd.org - dump.git/blob - CHANGES
Thanks for remote restore problem.
[dump.git] / CHANGES
1 $Id: CHANGES,v 1.144 2001/11/17 10:31:46 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 6. Add the -l option to restore to specify if, when doing a
25 remote restore, the file used to access the backup is a
26 regular file (the defaults being a tape drive). Restore needs
27 to know this information when reading a remote compressed
28 dump. Previously, this information was autodetected, but
29 the autodetection code fails (with ioctl: Inappropriate ioctl
30 for device) when using a non Linux remote box. Thanks to
31 many users and especially to Eros Albertazzi
32 <eros@lamel.bo.cnr.it> for reporting this.
33
34
35 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
36 ========================================================================
37
38 1. Fixed the permissions of a newly created QFA file by dump.
39
40 2. Cleaned up the source of dump (the external variables
41 definition was a complete mess, making possible to have
42 objects overlap).
43
44 3. Fixed restore to use the full tape volume path when doing
45 a compare (since it changes the working directory to the
46 filesystem being compared in the process).
47
48 4. Added the -q option to dump which makes dump abort
49 whenever it needs operator attention. It should be
50 easier to use dump in scripts with this option.
51
52 5. Detect the use of incompatible options to dump and
53 refuse them (like -a and -B options together).
54
55 6. Added bzip2 compression to dump/restore (use option -j level
56 to select it). Note that you will need a restore version
57 >= 0.4b24 in order to restore a bzip2 compressed dump.
58 The same warning as for the zlib compression applies:
59 the tape format of a bzip2 dump is not compatible with the
60 original BSD tape format.
61
62 7. Fixed a overflow problem in dump corrupting the dump when
63 very large files were encountered. Thanks to Vladimir Nadvornik
64 for the bug report.
65
66 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
67 kernel buffer/page cache before starting the dump, helping
68 a bit those who use dump on mounted filesystems. Thanks to
69 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
70 for this suggestion.
71
72 9. Updated the RPM spec file following the RedHat changes
73 (dynamically linked binaries now in /usr/sbin etc).
74
75 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
76 which enables restore to recognise multi volume compressed dumps
77 done on CD. Included his scripts for dump (which pipe the dump
78 output directly into cdrecord) and restore. There is now
79 possible to to backups to CD on the fly!
80
81 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
82 ===================================================================
83
84 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
85 Marcus Meissner (Caldera International Security Dept.).
86
87 2. Implement the Sun rmt extended protocol. Patch submitted
88 by Ian Gordon <iangordon@users.sourceforge.net>.
89
90 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
91 the linux kernel header. This ensures that dump always has the
92 latest version of this file. Patch submitted by
93 Andreas Dilger <adilger@turbolinux.com>.
94
95 4. Report any filesystem present in either /etc/fstab with a
96 positive passno or /etc/dumpdates in dump -w output.
97 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
98
99 5. Fixed the looping problem in dump introduced in the
100 previous version.
101
102 6. Changed the -B option of dump to limit the size of
103 _compressed_ output per volume if compression is on.
104 Patch contributed by Helmut Jarausch
105 <jarausch@igpm.rwth-aachen.de>. Note however that, since
106 it is impossible to predict the size of the compressed
107 data before doing the compression, the -B limit is a bit
108 conservative.
109
110 7. Fixed a bug in reading the operator typed file/tape path for
111 the next volume in restore.
112
113 8. Implemented a "-F script" option to restore which permits the
114 user to specify a script which will be launched at the
115 beginning of each tape, useful for automatic programming of
116 tape changers for example. See the restore man page for the
117 script parameters and return codes.
118
119 9. Small fix for the QFA routines provided by Uwe Gohlke
120 <uwe@ugsoft.de>, and some recommendations for QFA uses in
121 the man pages.
122
123 10. Fixed the multivolume restoring where making a mistake
124 to the 'Mount next tape' prompt caused several blocks to
125 be lost.
126
127 11. Enhanced the -e option of dump to take as a parameter a
128 comma separated list of inode numbers.
129
130 12. Added the -E option to dump which specify a file containing
131 inode numbers to exclude from the dump.
132
133 13. Fixed the compressed multi-volume dump + restore.
134
135 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
136 ==================================================================
137
138 1. Made dump report the number of blocks dumped per volume.
139 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
140
141 2. Fix a bug in dump where waiting too much at the 'change volumes'
142 question caused the volume to be rewritten. Thanks to
143 Graham Reed <greed@users.sourceforge.net> for reporting the
144 bug and providing a patch.
145
146 3. Added a compression option to dump, contributed by Jerry
147 Peters <gapeters@worldnet.att.net>.
148
149 WARNING: the compression works only on files/pipes and
150 drives supporting variable block size.
151
152 WARNING2: tapes created using the compression feature are
153 incompatible with the standard dump tape format, and a
154 version of restore >= 0.4b22 will be needed for extraction.
155
156 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
157 architectures. Thanks to Paul Slootman <paul@debian.org> for
158 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
159 it upstream.
160
161 5. Many cleanups (CPP defines, const char warnings, check of
162 ext2fs COMPAT flags, time_t cleanups, added libext2 version
163 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
164
165 6. Made --prefix option work in configure. All the install path
166 are now based on the configure parameters.
167
168 7. Added the Quick File Access mode in dump/restore, contributed
169 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
170 a file tape position for each inode, and this file is used by
171 restore (if called with parameter Q and the filename)
172 to directly position the tape at the file restore is currently
173 working on. This saves hours when restoring single files from
174 large backups, saves the tapes and the drive's head. Use
175 --enable-qfa option of configure to compile in the QFA support.
176
177 8. Added the possibility to dump several files and directories
178 in a single invocation of dump. Thanks to Uwe Gohlke
179 <uwe@ugsoft.de> for implementing this option.
180
181 9. Fixed the dumping and restoring of files with holes
182 (files ending with holes didn't get dumped or restored
183 correctly).
184
185 10. Fixed a socket descriptor leak in dump, which leaved opened
186 3 file descriptors per dump process (and there is one dump
187 process per tape).
188
189 11. Fixed dump large file system support, by explicit use of
190 open64/lseek64/etc functions (explicit use needed because
191 e2fsprogs libraries don't behave well when compiled with
192 FILE_OFFSET_BITS=64).
193
194 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
195 ======================================================================
196
197 1. Fixed some bugs in the dump -w|-W logic introduced by
198 the previous version. Thanks to Andreas Dilger
199 <adilger@turbolinux.com> for his help on this one.
200
201 2. Fixed again a compilation problem when using old e2fs
202 headers (filesystem label related). Thanks to many users
203 who reported this stupid error.
204
205 3. Fixed a build problem on old lib5 based systems dealing with
206 _PATH_MNTTAB being undefined. Thanks to John Adams
207 <johna@onevista.com> for reporting the bug.
208
209 4. Improved the error detection in restore when creating
210 the temporary files in TMPDIR. Now restore will corectly
211 report a 'no space left on device' error instead of
212 strange errors which could imply an archive corruption.
213 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
214 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
215
216 5. Added the throughput information in dump's progression
217 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
218 for the patch.
219
220 6. Use libext2fs's inode scanning routines, which are particularly
221 robust when dealing with errors (especially when having some
222 bad blocks on the filesystem). Many thanks to Colin
223 <colin@horizon.com> for reporting the bug, investigating
224 the issues, finding a workaround, writing the patch and
225 fully testing it... (of course, if this patch will break
226 anything, he is to blame for :-)).
227
228 7. Made dump and restore LFS aware. Dump can dump a filesystem
229 containing large files, generate a large file on output and
230 restore can restore them. This feature is not enabled by
231 default, use --enable-largefile option of configure to enable
232 it (you need to have a LFS aware glibc though). Thanks to
233 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
234 and to Theodore T'so <tytso@valinux.com> for his always
235 useful thoughts.
236
237 8. Made dump ask upon a tape write error if it should rewrite
238 the current volume (assume this is a bad tape) or if it should
239 assume an end-of-tape condition (useful for tape drives which
240 behaves badly at the end of the tape). Thanks to Andreas
241 Dilger <adilger@turbolinux.com> for the suggestion.
242
243 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
244 =======================================================================
245
246 1. Fixed a small compilation problem due to a change
247 in the definintion of the struct sigaction in
248 glibc 2.0 / libc5. Thanks to Gunther Schlegel
249 <schlegel@riege.de> for reporting the bug and to
250 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
251 a fix.
252
253 2. Modified the label and uuid analysis in order to be
254 self-contained (independant of kernel/libc headers). This
255 should fix the compile with older kernel/libc headers and
256 will preserve the functionality. Thanks to Bernhard Erdmann
257 <bernhard.erdmann@gmx.de> for reporting the bug.
258
259 3. The 'exclude inode' option, if applied to a directory
260 inode, excludes now correctly all the leaves of this
261 directory. Thanks to John R. Dennison
262 <gerdesas@users.sourceforge.net> for reporting the bug.
263
264 4. Fixed the '-e' option to disable the possibility
265 to exclude the root inode (which causes the dumps to
266 be unreadable by restore). Prevented array overflow
267 when multiple -e options are used.
268
269 5. Fixed dump to correctly interpret a filesystem argument
270 which is a mountpoint and it is not an absolute path
271 (as specified in the fstab). Thanks to Bernhard R. Erdmann
272 <be@berdmann.de> for reporting the bug.
273
274 6. Made dump able to backup files larger than 2 GB. Note that
275 dump still doesn't cope with files larger than 4 GB.
276
277 7. Restore the real uid and gid before invoking an external
278 RSH program (big hole when dump or restore is suid root!).
279
280 8. Get the values for the default tape device and the location
281 of fstab file from the system headers. Thanks to
282 Andreas Dilger <adilger@turbolinux.com> for the patch.
283
284 9. Made dump -w|-W report all recognized filesystems
285 present in either /etc/fstab or /etc/dumpdates, and present
286 the list in the same order as in fstab file. Thanks
287 to Andreas Dilger <adilger@turbolinux.com> for the patch.
288
289 10. Made dump's -a (automatic end of tape detection) the
290 default. Specifying one of -B, -c, -d or -s options will
291 override the EOM detection. Thanks to Andreas Dilger
292 <adilger@turbolinux.com> for the patch.
293
294 11. Save the ext2 filesystem label into the dump volume label.
295 Specifying a volume label on the command line (-L option)
296 will override this feature. Thanks to Andreas Dilger
297 <adilger@turbolinux.com> for the patch.
298
299 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
300 =====================================================================
301
302 1. Fixed the signal handling in dump (which I broke in 0.4b17)
303 which was causing several strange dump failures (dump
304 hanged or segmentation faults etc).
305
306 2. Specified the default blocksize in the dump man page.
307
308 3. Changed two info messages of restore to be written on stdout
309 instead of stderr in order to leave stderr only for errors
310 or warnings. Thanks to Stephen Atwell
311 <satwell@urbana.css.mot.com> for the suggestion.
312
313 4. Corrected an off by one calculation which prevented
314 dumping a directory having a 1 character name length.
315 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
316 for reporting the bug.
317
318 5. Reinforce dump in order to allow only level 0 and no
319 -u option when dumping a subdirectory, and document
320 this in the man page. Thanks to Bernhard Erdmann
321 <bernhard.erdmann@gmx.de> for reporting the bug.
322
323 6. Fixed a small harmless bug in restore which caused
324 in some conditions hard links to be restored several
325 times (therefore generation some warning messages).
326 Thanks to Kenneth Porter <shiva@well.com> for
327 reporting the bug.
328
329 7. Updated the RPM spec file to the latest RedHat version,
330 providing FHS packaging and other cosmetic changes.
331 You will need RPM version >= 3.0.5 in order to build the RPM.
332
333 8. Updated the configure script to check for libtermcap
334 before checking for libreadline, because we need this
335 library in order to compile the readline support.
336
337 9. Made dump understand the LABEL= and UUID= notation
338 both in /etc/fstab and on the command line. Note that
339 you will need the /proc filesystem in order to use
340 these notations. Thanks to Erik Troan <ewt@redhat.com>
341 for providing the patch.
342
343 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
344 ===================================================================
345
346 1. Fixed a potential buffer overflow in restore. Thanks
347 to Stan Bubrouski <satan@fastdial.net> for reporting
348 the bug.
349
350 2. Fixed a readline-related bug which prevented
351 'cat DUMPFILE | restore -i -f -' from working. Thanks
352 to Charles Karney <karney@users.sourceforge.net>
353 for the bug report.
354
355 3. Changed a few "panic" into "exit", causing restore to
356 be more stable against some attacks (like the last one
357 reported on Bugtraq, although the last version of restore
358 was not vulnerable - just dumped core). Thanks to
359 Andreas Hasenack <andreas@conectiva.com.br> for reporting
360 the bugs to me.
361
362 4. Removed the suid-root bit on dump and restore in the
363 default build (and generated RPMs). It should be safer
364 now. Users who need the suid-root capabilities in order
365 to make network backups should read first the man page
366 of dump and enable the suid bit manually.
367
368 5. Added -ltermcap to the compile parameters for restore
369 when using readline support, in order to make the compile
370 process work on some systems (SuSE ?). Thanks to
371 Patrik Schindler <poc@pocnet.net> for reporting the bug.
372
373 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
374 ====================================================================
375
376 1. The -F script is called now *only* at the end of a tape,
377 not at the end of the dump. Thanks to Christian Haul
378 <haul@informatik.tu-darmstadt.de> for the bug report.
379
380 Normally, the device name and the current volume number
381 are passed on the command line of the script. However,
382 if you want the old style script (with no arguments
383 passed on the command line), you can enable it in
384 configure with the --enable-oldstylefscript.
385
386 2. Use posix signal handling to preserve dump functionality
387 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
388 the patch.
389
390 3. Made the exit code of restore in case of a 'C'ompare
391 command reflect the result of the compare. An exit status
392 of 0 means the dump archive is correct, 1 means tape errors,
393 2 means that some files were modified. Thanks to Kenneth Porter
394 <shiva@well.com> for the suggestion.
395
396 4. Made (finally) quotes work in interactive restore.
397
398 5. Small fixes in order to allow dump to compile with a
399 really old e2fsprogs version. Thanks to Ian Zimmerman
400 <itz@speakeasy.org> for the bug report.
401
402 6. Add GNU readline capabilities to interactive restore.
403 Use configure's --enable-readline flag to enable this feature.
404 Thanks to Patrik Schindler <poc@pocnet.net> for the
405 suggestion.
406
407 7. Do the compare on the fly in restore 'C' mode (this will
408 allow not to exhaust the available /tmp space when
409 ccmparing large files). Thanks to Kenneth Porter
410 <shiva@well.com> for the suggestion.
411
412 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
413 ====================================================================
414
415 1. Fixed some several duplicate 'DUMP: DUMP:' in the
416 output of dump.
417
418 2. Corrected the estimation of blocks to dump. Note that
419 this is still wrong for large directory inodes, due
420 to the size of a BSD directory entry on the tape
421 (estimation is based on the size of the ext2 directory,
422 which is smaller because it doesn't contain the null
423 character at the end).
424
425 3. Made dump report the total number of MB written to
426 tape at the end of run. Thanks to W. Reilly Cooley
427 <wcooley@nakedape.cc> for the patch.
428
429 4. Added the -X option to restore giving the possibility
430 to read the names of the files to be extracted/listed
431 from a text file (in addition of the command line).
432 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
433 patch.
434
435 5. Added the device name and the current volume number
436 as arguments to the end of tape script (dump -F option).
437
438 6. Made the multi-volume dump work again (after having
439 broken it in 0.4b15).
440
441 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
442 ===================================================================
443
444 1. Added a prompt command in interactive restore mode. Thanks
445 to Andreas Dilger <adilger@home.com> for the patch.
446
447 2. Fixed a buffer overflow problem in dump (caused by
448 not checking the size of the filesystem parameter).
449 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
450 reporting this on Bugtraq (and to several dump users
451 who forwarded me his mail).
452
453 3. Added the '-F script' option to dump in order to
454 launch a script at the end of each tape (to be used
455 with a tape changer, or to notify the sysadmin by
456 pager etc.).
457
458 4. Fixed a bug in restore compare code caused by the changes
459 I made in 0.4b14.
460
461 5. Fixed the treatment of options using the old BSD syntax
462 in both dump and restore.
463
464 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
465 =======================================================================
466
467 1. Fixed a bug in dump which may have caused invalid deleted
468 directories to be dumped out if they were deleted after the
469 mapping pass. This could occure on active filesystem and lead
470 to heap corruption (causing dump malfunction in many possible ways).
471 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
472 bug and submitting the patch.
473
474 2. Fixed the handling of the filesystem parameter in dump. A
475 '/mnt/disk' parameter caused the disk contents to be dumped,
476 but a '/mnt/disk/' parameter caused the mountpoint directory
477 to be dumped (generally an empty directory).
478
479 3. Improved the output of dump in order to tell which directory
480 it is currently dumping (when dumping a subtree).
481
482 4. Added the '-e' exclude inode option to dump. Thanks to
483 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
484
485 5. Added a REPORTING-BUGS file in order to provide a guide
486 on how to correctly report a bug in dump/restore.
487
488 6. Fixed a restore problem when restoring a hard link to an inode
489 having the immutable or append-only attribute set. Thanks to
490 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
491
492 7. Fixed a compatibility problem between dumps made on little
493 endian machines (the format was correct) and big endian
494 machines (the format was incorrect). This fix break the
495 compatibility with the older dumps made on big endian
496 machines (sparc, m86k, ppc etc). For the first time in
497 linux dump's history (I believe), the dumps made by *BSD,
498 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
499 so interchangeable. Thanks to Rob Cermak
500 <cermak@ahab.rutgers.edu> for submitting the bug and
501 helping me test the fix.
502
503 8. Fixed the way dump reports the remaining percent/time, if
504 the number of blocks actually dumped exceeds the estimated
505 number of blocks. Thanks to Jean-Paul van der Jagt
506 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
507
508 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
509 ======================================================================
510
511 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
512 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
513
514 2. Removed the requirement to build the RPM as root from the
515 spec file. Thanks to Christian Weisgerber
516 <naddy@mips.rhein-neckar.de> for submitting this.
517
518 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
519 causing dump to block when dumping really huge filesystems.
520 Many thanks to Patrik Schindler <poc@pocnet.net> for
521 helping me find this bug.
522
523 4. Fixed the treatment for an interrupt signal when dump access
524 the remote tape through RSH. Thanks to Christian Weisgerber
525 <naddy@mips.rhein-neckar.de> for providing the patch.
526
527 5. Fixed a bug which was causing dump/restore to display
528 garbage characters instead of the remote host name.
529
530 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
531 =====================================================================
532
533 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
534 <kukuk@suse.de> for submitting the patch.
535
536 2. Fix for the exit code when using the size estimate option of
537 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
538 submitting the patch.
539
540 3. Handle EINTR in atomical reads/writes in dump, which was causing
541 dump to fail on some systems. Thanks to Eric Jergensen
542 <eric@dvns.com> for reporting the bug and submitting the patch.
543
544 4. Handle more than 16 characters for the device names in dumpdates.
545 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
546 tracking down the problem and proposing the solution.
547
548 5. Fixed a bug in dump which prevented the creation of the
549 dumpdates file when doing a 0-level dump without already
550 having a dumpdates file. Thanks to Patrik Schindler
551 <poc@pocnet.net> for reporting the bug.
552
553 6. Changed the way dump 'S' flag reports the size estimate
554 from number of blocks into bytes (making it compatible
555 with the Solaris version, and simplifying things for
556 amanda users). Thanks to Jason L Tibbitts III
557 <tibbs@math.uh.edu> for reporting the bug.
558
559 7. Fixed a compatibility problem in linux/alpha dump tape format.
560 Now the linux/alpha dump are (again) compatible with the
561 other dump formats. But this breaks compatibility with
562 older dumps made on alpha. Thanks to Mike Tibor
563 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
564
565 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
566 ======================================================================
567
568 1. Added a '--enable-kerberos' to configure.
569
570 2. Added a 'S' option to dump which determines the amount of space
571 that is needed to perform the dump without actually doing it, similar
572 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
573 <cermak@ahab.rutgers.edu>.
574
575 3. Added a 'M' multi-volume option to dump and restore which enables
576 dumping to multiple files (useful when dumping to an ext2
577 partition to make several dump files in order to bypass the 2GB
578 file size limitation). The 'f' argument is treated as a prefix and
579 the output files will be named <prefix>001, <prefix>002 etc. With
580 the 'M' flag, restore automatically selects the right file without
581 asking to enter a new tape each time.
582
583 4. Fixed a memory leak which was causing dump to grow very big
584 (270MB when dumping a 10GB filesystem...). Thanks to Jason
585 Fearon <jasonf@netrider.org.au> for reporting the bug.
586
587 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
588 ======================================================================
589
590 1. Make configure test if the system glob routines support
591 extended syntax (ALTDIRFUNC). If not, use the internal glob
592 routines instead of system ones. Thanks to Bernhard Sadlowski
593 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
594 and helping me resolve this and other minor libc5 compiling
595 glitches.
596
597 2. Fix a problem when dumping a ext2fs with the 'filetype'
598 feature enabled. Thanks to Patrick J. LoPresti
599 <patl@cag.lcs.mit.edu> for reporting the bug and to
600 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
601
602 3. Made the nodump flag work on directories. A directory which
603 has the nodump flag gets never dumped, regardless of its
604 contents.
605
606 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
607 which allows dump on an active ext3 filesystem. However, this
608 is a "quick and dirty" patch which enables backup of an ext3
609 filesystem through the ext2 compatibility (by ignoring the
610 NEEDS_RECOVERY bit). The journal file is not recognized and
611 it is dumped (it should not).
612
613 5. Test the superblock compatibility flags when dumping, in order
614 to be sure that we know how to deal with specific features.
615
616 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
617 ====================================================================
618
619 1. Use lchown instead of chown, fixing a possible security problem
620 when restoring symlinks (a malicious user could use this
621 to deliberately corrupt the ownership of important system files).
622 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
623 this and providing the patch.
624
625 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
626 ====================================================================
627
628 1. Put dump sources under CVS, added Id tags in all files so
629 one can use 'ident' on binary files.
630
631 2. Added the dump/restore version in the usage text so one can
632 easily verify the version he is using.
633
634 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
635 a va_start/va_end problem on linux-ppc (always call va_start
636 va_end in pairs each time we use a vararg function).
637
638 4. Added again the DT_* constants because old libc does not
639 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
640 for submitting the bug report.
641
642 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
643 this should enable dumping big (huge) filesystems.
644
645 6. Added the RSH environment variable in order to be able to
646 use a rsh replacement like ssh when doing remote backups (and
647 bypass the security limitations of rcmd). Now you can do remote
648 backups without being root (or making dump setuid root).
649
650 7. Modified again the way dumpdates works. For incremental dumps,
651 we need to read dumpdates even if we are not using 'u' option.
652 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
653 this should work.
654
655 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
656 ===================================================================
657
658 1. Removed the 'k' flag from the restore 'about' text if kerberos
659 was not compiled in.
660
661 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
662 to them (fsetflags takes a char*, setflags an open fd!).
663
664 3. (f)setflags is called only if the flags aren't empty. If the
665 file is a special file, a warning is printed, because changing
666 flags implies opening the device. Normally, a special file
667 should not have any flag... (Debian bug #29775, patch provided
668 by Abhijit Dasgupta <abhijit@ans.net>).
669
670 4. Made possible to dump a file system not mentioned in /etc/fstab.
671 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
672
673 5. Changed the default behaviour to not create dumpdates
674 unless 'u' option is specified. Removed the old "debian-patch"
675 which provided the same thing. (Debian bug #38136, #33818).
676
677 6. Removed all those dump*announce, since they were getting old...
678
679 7. Added warning messages if dumpdates does not exist and
680 when an inferior level dump does not exist (except for a level 0
681 dump).
682
683 8. Debugged the glob calls in interactive mode: restore used a
684 dirent struct which was different from the /usr/include/dirent.h
685 one (this used to work, is it a glibc2 change?), so none of the
686 compat glob (which used /usr/include/dirent.h) or the system glob
687 worked. Restore use now the system dirent (and the system
688 DT_* constants), which are compatible with BSD ones.
689
690 9. Added a configure flag (--with-dumpdatespath) to specify
691 the location of dumpdates file. By default, it is
692 /etc/dumpdates.
693
694 10. Added the "AUTHOR" and "AVAILABILITY" sections and
695 included the current date/version in man pages.
696
697 11. Corrected the estimation of remaining time when
698 the operator doesn't change the tapes quickly enough. This
699 was an old bug, I thought I corrected it, and discovered
700 that in fact it was corrected in two different places, so
701 the results canceled each other...
702
703 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
704 ===================================================================
705
706 1. Integrated multiple patches from RedHat, Debian and SuSE:
707
708 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
709 isn't specified.
710 - several fixes in the man pages.
711 - update the default tape device to /dev/st0.
712 - many updates for Linux Alpha (byte ordering, size_t etc).
713 - buffer overruns.
714 - use environment variable for TMPDIR (instead of /tmp).
715 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
716 - workaround egcs bug (RedHat bugs #4281 and #2989).
717 - wire $(OPT) throughout Makefile's.
718
719 2. Upgrade the dump revision to 1, making possible to dump filesystems
720 made with e2fsprogs-1.15 or newer. Nothing seems to break...
721
722 3. Fix some compile warnings, prototype all functions.
723
724 4. Use glibc err/glob instead of internal compatibility
725 routines (only if available).
726
727 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
728 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
729
730 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
731 ======================================================================
732
733 1. Integrated the changes from FreeBSD-3.1-RELEASE
734 (mostly bug fixes, buffer overruns, dump has now an "automatic
735 tape length calculation" flag, dump/restore can use kerberos now
736 (this is NOT tested), use environment variables for TAPE and
737 RMT etc.).
738
739 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
740
741 3. Corrected a bug in restore when using 'C' option with multi-volumes
742 tapes (files splited accros two tapes give "size changed" errors
743 when comparing).
744
745 4. Corrected the long standing bug when dumping multiple tapes.
746 This works for me, needs further testing.
747
748 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
749 ====================================================================
750
751 1. Dump now runs correctly on kernels 2.1.x
752 Fix made by Gerald Peters <gapeters@worldnet.att.net>
753
754 Changes between versions 0.4b2 and 0.4b3
755 ========================================
756
757 1. Use realpath() if available
758
759 2. Report statistics
760
761 Changes between versions 0.4b1 and 0.4b2
762 ========================================
763
764 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
765 the patch)
766
767 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
768
769 3. Fixed configure.in to correctly handle --enable-rmt
770
771 Changes between versions 0.3 and 0.4b1
772 ======================================
773
774 1. Integrated the changes from 4.4BSD-Lite2
775
776 2. Integrated the patches from Debian and Red Hat
777
778 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
779
780 4. Changed dump to use the Ext2fs library to get block addresses. This
781 should solve the endianness problem on SparcLinux.
782
783 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
784 distribution's one) to use autoconf
785
786 6. Fixed a few minor bugs
787
788 Changes between versions 0.2e and 0.2f
789 ======================================
790
791 1. Added the creation of named pipes (fifos) in restore.
792
793 2. Added the -N flag in the restore manual page.
794
795 3. Added the file kernel.patch which contains the llseek() optimization
796 patch for 1.2.x kernels.
797
798 4. Fixed a bug in the restoration of symbolic links: owner and group were
799 not restored.
800
801 5. Integrated some changes from FreeBSD 2.2.
802
803 6. Added a call to ftruncate() after restoring each file to restore
804 correctly files ending by a hole.
805
806 Changes between versions 0.2d and 0.2e
807 ======================================
808
809 1. Fixed a bug in the "set owner/mode" process. Every file was restored
810 with owner = root (0) and group = root/wheel/whatever (0).
811
812 Changes between versions 0.2c and 0.2d
813 ======================================
814
815 1. Dump is now able to backup 2GB+ filesystems.
816
817 2. Dump and restore can now be linked as static binaries.
818
819 Changes between versions 0.2b and 0.2c
820 ======================================
821
822 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
823
824 Changes between versions 0.2a and 0.2b
825 ======================================
826
827 1. Really fixed the bug that I should have corrected in 0.2a.
828
829 2. Enabled optimization again.
830
831 Changes between versions 0.2 and 0.2a
832 =====================================
833
834 1. Disabled the optimization during compilation.
835
836 Changes between versions 0.1 and 0.2
837 ====================================
838
839 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
840 the fs_type field (actually, I modified the file fstab.c to make it
841 use the mntent functions).
842
843 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
844 made by dump should be readable by the BSD restore and backups made
845 by the BSD dump should be readable by restore. Unfortunately, this
846 means that the dump format has changed between version 0.1 and version
847 0.2 :-(
848
849 3. Dump is now able to backup a subtree, it is no longer limited to whole
850 filesystems like the BSD version.
851
852 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
853 than 2 GB.
854
855 Changes between versions 0.0 and 0.1
856 ====================================
857
858 1. Now create links rdump and rrestore during the `make install' step.
859
860 2. Linux port specific bugs added to the manual pages
861
862 3. Incorrect estimation of the number of tapes blocks fixed when doing
863 incremental backups.
864
865 4. Better ls-like format in restore in interactive mode.