]> git.wh0rd.org - dump.git/blob - CHANGES
SIGXFSZ fixes.
[dump.git] / CHANGES
1 $Id: CHANGES,v 1.154 2002/01/10 09:02:54 stelian Exp $
2
3 Changes between versions 0.4b26 and 0.4b27 (released ???????????????)
4 =====================================================================
5
6 1. Fixed behaviour of dump when exceeding resource limits
7 (SIGXFSZ treatment).
8
9 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
10 =====================================================================
11
12 1. Added a set of backup scripts from Eugenio Diaz
13 <getnito@yahoo.com> in the examples section. It features
14 automatic (cron based) full and incremental dumping of
15 several filesystems on a separate filesystem.
16
17 2. Fixed a off-by-one miscalculation which disabled dumping
18 a one letter subdirectory. Thanks to Chris Siebenmann
19 <cks@utcc.utoronto.ca> for reporting the bug.
20
21 3. Fixed several restore bugs occuring when trying to
22 restore some missing files on the tape. Thanks to Chris
23 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
24
25 4. Fixed --with-ldopts configure argument passing, installing from
26 a separate object directory, makefile cleanups contributed
27 by <splite@purdue.edu>.
28
29 5. Fix a bug which could caused, in some conditions, the highest
30 number inode of a filesystem, to not be dumped. Many thanks
31 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
32 chase this bug.
33
34 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
35 =======================================================================
36
37 1. Added a mini howto from Patrick Walsh in the examples
38 directory.
39
40 2. Minor man pages syntax corrections. Thanks to
41 Chris Danis <screechco@home.com> for reporting the bugs.
42
43 3. Added a script from David B. Peterson <dave@toppledwagon.com>
44 to the examples section. It features dumping several
45 filesystems to a remote tape drive upon ssh.
46
47 4. Added a patch provided by Richard Jones <rich@annexia.org>
48 which allows BRADEMAX (number of read errors tolerated by
49 dump) to be adjusted using the -I option.
50
51 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
52 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
53 patch.
54
55 6. Add the -l option to restore to specify if, when doing a
56 remote restore, the file used to access the backup is a
57 regular file (the defaults being a tape drive). Restore needs
58 to know this information when reading a remote compressed
59 dump. Previously, this information was autodetected, but
60 the autodetection code fails (with ioctl: Inappropriate ioctl
61 for device) when using a non Linux remote box. Thanks to
62 many users and especially to Eros Albertazzi
63 <eros@lamel.bo.cnr.it> for reporting this.
64
65 7. Found a workaroung for the dump deadlock problem (3 childs
66 stuck in pause(), father in read()). The workaround seems
67 to work for me and several beta-testers. If it doesn't work
68 for you, please report back.
69
70 8. Updated the RPM spec file (BuildPrereq, URL etc).
71
72 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
73 ========================================================================
74
75 1. Fixed the permissions of a newly created QFA file by dump.
76
77 2. Cleaned up the source of dump (the external variables
78 definition was a complete mess, making possible to have
79 objects overlap).
80
81 3. Fixed restore to use the full tape volume path when doing
82 a compare (since it changes the working directory to the
83 filesystem being compared in the process).
84
85 4. Added the -q option to dump which makes dump abort
86 whenever it needs operator attention. It should be
87 easier to use dump in scripts with this option.
88
89 5. Detect the use of incompatible options to dump and
90 refuse them (like -a and -B options together).
91
92 6. Added bzip2 compression to dump/restore (use option -j level
93 to select it). Note that you will need a restore version
94 >= 0.4b24 in order to restore a bzip2 compressed dump.
95 The same warning as for the zlib compression applies:
96 the tape format of a bzip2 dump is not compatible with the
97 original BSD tape format.
98
99 7. Fixed a overflow problem in dump corrupting the dump when
100 very large files were encountered. Thanks to Vladimir Nadvornik
101 for the bug report.
102
103 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
104 kernel buffer/page cache before starting the dump, helping
105 a bit those who use dump on mounted filesystems. Thanks to
106 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
107 for this suggestion.
108
109 9. Updated the RPM spec file following the RedHat changes
110 (dynamically linked binaries now in /usr/sbin etc).
111
112 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
113 which enables restore to recognise multi volume compressed dumps
114 done on CD. Included his scripts for dump (which pipe the dump
115 output directly into cdrecord) and restore. There is now
116 possible to to backups to CD on the fly!
117
118 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
119 ===================================================================
120
121 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
122 Marcus Meissner (Caldera International Security Dept.).
123
124 2. Implement the Sun rmt extended protocol. Patch submitted
125 by Ian Gordon <iangordon@users.sourceforge.net>.
126
127 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
128 the linux kernel header. This ensures that dump always has the
129 latest version of this file. Patch submitted by
130 Andreas Dilger <adilger@turbolinux.com>.
131
132 4. Report any filesystem present in either /etc/fstab with a
133 positive passno or /etc/dumpdates in dump -w output.
134 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
135
136 5. Fixed the looping problem in dump introduced in the
137 previous version.
138
139 6. Changed the -B option of dump to limit the size of
140 _compressed_ output per volume if compression is on.
141 Patch contributed by Helmut Jarausch
142 <jarausch@igpm.rwth-aachen.de>. Note however that, since
143 it is impossible to predict the size of the compressed
144 data before doing the compression, the -B limit is a bit
145 conservative.
146
147 7. Fixed a bug in reading the operator typed file/tape path for
148 the next volume in restore.
149
150 8. Implemented a "-F script" option to restore which permits the
151 user to specify a script which will be launched at the
152 beginning of each tape, useful for automatic programming of
153 tape changers for example. See the restore man page for the
154 script parameters and return codes.
155
156 9. Small fix for the QFA routines provided by Uwe Gohlke
157 <uwe@ugsoft.de>, and some recommendations for QFA uses in
158 the man pages.
159
160 10. Fixed the multivolume restoring where making a mistake
161 to the 'Mount next tape' prompt caused several blocks to
162 be lost.
163
164 11. Enhanced the -e option of dump to take as a parameter a
165 comma separated list of inode numbers.
166
167 12. Added the -E option to dump which specify a file containing
168 inode numbers to exclude from the dump.
169
170 13. Fixed the compressed multi-volume dump + restore.
171
172 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
173 ==================================================================
174
175 1. Made dump report the number of blocks dumped per volume.
176 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
177
178 2. Fix a bug in dump where waiting too much at the 'change volumes'
179 question caused the volume to be rewritten. Thanks to
180 Graham Reed <greed@users.sourceforge.net> for reporting the
181 bug and providing a patch.
182
183 3. Added a compression option to dump, contributed by Jerry
184 Peters <gapeters@worldnet.att.net>.
185
186 WARNING: the compression works only on files/pipes and
187 drives supporting variable block size.
188
189 WARNING2: tapes created using the compression feature are
190 incompatible with the standard dump tape format, and a
191 version of restore >= 0.4b22 will be needed for extraction.
192
193 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
194 architectures. Thanks to Paul Slootman <paul@debian.org> for
195 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
196 it upstream.
197
198 5. Many cleanups (CPP defines, const char warnings, check of
199 ext2fs COMPAT flags, time_t cleanups, added libext2 version
200 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
201
202 6. Made --prefix option work in configure. All the install path
203 are now based on the configure parameters.
204
205 7. Added the Quick File Access mode in dump/restore, contributed
206 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
207 a file tape position for each inode, and this file is used by
208 restore (if called with parameter Q and the filename)
209 to directly position the tape at the file restore is currently
210 working on. This saves hours when restoring single files from
211 large backups, saves the tapes and the drive's head. Use
212 --enable-qfa option of configure to compile in the QFA support.
213
214 8. Added the possibility to dump several files and directories
215 in a single invocation of dump. Thanks to Uwe Gohlke
216 <uwe@ugsoft.de> for implementing this option.
217
218 9. Fixed the dumping and restoring of files with holes
219 (files ending with holes didn't get dumped or restored
220 correctly).
221
222 10. Fixed a socket descriptor leak in dump, which leaved opened
223 3 file descriptors per dump process (and there is one dump
224 process per tape).
225
226 11. Fixed dump large file system support, by explicit use of
227 open64/lseek64/etc functions (explicit use needed because
228 e2fsprogs libraries don't behave well when compiled with
229 FILE_OFFSET_BITS=64).
230
231 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
232 ======================================================================
233
234 1. Fixed some bugs in the dump -w|-W logic introduced by
235 the previous version. Thanks to Andreas Dilger
236 <adilger@turbolinux.com> for his help on this one.
237
238 2. Fixed again a compilation problem when using old e2fs
239 headers (filesystem label related). Thanks to many users
240 who reported this stupid error.
241
242 3. Fixed a build problem on old lib5 based systems dealing with
243 _PATH_MNTTAB being undefined. Thanks to John Adams
244 <johna@onevista.com> for reporting the bug.
245
246 4. Improved the error detection in restore when creating
247 the temporary files in TMPDIR. Now restore will corectly
248 report a 'no space left on device' error instead of
249 strange errors which could imply an archive corruption.
250 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
251 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
252
253 5. Added the throughput information in dump's progression
254 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
255 for the patch.
256
257 6. Use libext2fs's inode scanning routines, which are particularly
258 robust when dealing with errors (especially when having some
259 bad blocks on the filesystem). Many thanks to Colin
260 <colin@horizon.com> for reporting the bug, investigating
261 the issues, finding a workaround, writing the patch and
262 fully testing it... (of course, if this patch will break
263 anything, he is to blame for :-)).
264
265 7. Made dump and restore LFS aware. Dump can dump a filesystem
266 containing large files, generate a large file on output and
267 restore can restore them. This feature is not enabled by
268 default, use --enable-largefile option of configure to enable
269 it (you need to have a LFS aware glibc though). Thanks to
270 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
271 and to Theodore T'so <tytso@valinux.com> for his always
272 useful thoughts.
273
274 8. Made dump ask upon a tape write error if it should rewrite
275 the current volume (assume this is a bad tape) or if it should
276 assume an end-of-tape condition (useful for tape drives which
277 behaves badly at the end of the tape). Thanks to Andreas
278 Dilger <adilger@turbolinux.com> for the suggestion.
279
280 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
281 =======================================================================
282
283 1. Fixed a small compilation problem due to a change
284 in the definintion of the struct sigaction in
285 glibc 2.0 / libc5. Thanks to Gunther Schlegel
286 <schlegel@riege.de> for reporting the bug and to
287 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
288 a fix.
289
290 2. Modified the label and uuid analysis in order to be
291 self-contained (independant of kernel/libc headers). This
292 should fix the compile with older kernel/libc headers and
293 will preserve the functionality. Thanks to Bernhard Erdmann
294 <bernhard.erdmann@gmx.de> for reporting the bug.
295
296 3. The 'exclude inode' option, if applied to a directory
297 inode, excludes now correctly all the leaves of this
298 directory. Thanks to John R. Dennison
299 <gerdesas@users.sourceforge.net> for reporting the bug.
300
301 4. Fixed the '-e' option to disable the possibility
302 to exclude the root inode (which causes the dumps to
303 be unreadable by restore). Prevented array overflow
304 when multiple -e options are used.
305
306 5. Fixed dump to correctly interpret a filesystem argument
307 which is a mountpoint and it is not an absolute path
308 (as specified in the fstab). Thanks to Bernhard R. Erdmann
309 <be@berdmann.de> for reporting the bug.
310
311 6. Made dump able to backup files larger than 2 GB. Note that
312 dump still doesn't cope with files larger than 4 GB.
313
314 7. Restore the real uid and gid before invoking an external
315 RSH program (big hole when dump or restore is suid root!).
316
317 8. Get the values for the default tape device and the location
318 of fstab file from the system headers. Thanks to
319 Andreas Dilger <adilger@turbolinux.com> for the patch.
320
321 9. Made dump -w|-W report all recognized filesystems
322 present in either /etc/fstab or /etc/dumpdates, and present
323 the list in the same order as in fstab file. Thanks
324 to Andreas Dilger <adilger@turbolinux.com> for the patch.
325
326 10. Made dump's -a (automatic end of tape detection) the
327 default. Specifying one of -B, -c, -d or -s options will
328 override the EOM detection. Thanks to Andreas Dilger
329 <adilger@turbolinux.com> for the patch.
330
331 11. Save the ext2 filesystem label into the dump volume label.
332 Specifying a volume label on the command line (-L option)
333 will override this feature. Thanks to Andreas Dilger
334 <adilger@turbolinux.com> for the patch.
335
336 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
337 =====================================================================
338
339 1. Fixed the signal handling in dump (which I broke in 0.4b17)
340 which was causing several strange dump failures (dump
341 hanged or segmentation faults etc).
342
343 2. Specified the default blocksize in the dump man page.
344
345 3. Changed two info messages of restore to be written on stdout
346 instead of stderr in order to leave stderr only for errors
347 or warnings. Thanks to Stephen Atwell
348 <satwell@urbana.css.mot.com> for the suggestion.
349
350 4. Corrected an off by one calculation which prevented
351 dumping a directory having a 1 character name length.
352 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
353 for reporting the bug.
354
355 5. Reinforce dump in order to allow only level 0 and no
356 -u option when dumping a subdirectory, and document
357 this in the man page. Thanks to Bernhard Erdmann
358 <bernhard.erdmann@gmx.de> for reporting the bug.
359
360 6. Fixed a small harmless bug in restore which caused
361 in some conditions hard links to be restored several
362 times (therefore generation some warning messages).
363 Thanks to Kenneth Porter <shiva@well.com> for
364 reporting the bug.
365
366 7. Updated the RPM spec file to the latest RedHat version,
367 providing FHS packaging and other cosmetic changes.
368 You will need RPM version >= 3.0.5 in order to build the RPM.
369
370 8. Updated the configure script to check for libtermcap
371 before checking for libreadline, because we need this
372 library in order to compile the readline support.
373
374 9. Made dump understand the LABEL= and UUID= notation
375 both in /etc/fstab and on the command line. Note that
376 you will need the /proc filesystem in order to use
377 these notations. Thanks to Erik Troan <ewt@redhat.com>
378 for providing the patch.
379
380 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
381 ===================================================================
382
383 1. Fixed a potential buffer overflow in restore. Thanks
384 to Stan Bubrouski <satan@fastdial.net> for reporting
385 the bug.
386
387 2. Fixed a readline-related bug which prevented
388 'cat DUMPFILE | restore -i -f -' from working. Thanks
389 to Charles Karney <karney@users.sourceforge.net>
390 for the bug report.
391
392 3. Changed a few "panic" into "exit", causing restore to
393 be more stable against some attacks (like the last one
394 reported on Bugtraq, although the last version of restore
395 was not vulnerable - just dumped core). Thanks to
396 Andreas Hasenack <andreas@conectiva.com.br> for reporting
397 the bugs to me.
398
399 4. Removed the suid-root bit on dump and restore in the
400 default build (and generated RPMs). It should be safer
401 now. Users who need the suid-root capabilities in order
402 to make network backups should read first the man page
403 of dump and enable the suid bit manually.
404
405 5. Added -ltermcap to the compile parameters for restore
406 when using readline support, in order to make the compile
407 process work on some systems (SuSE ?). Thanks to
408 Patrik Schindler <poc@pocnet.net> for reporting the bug.
409
410 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
411 ====================================================================
412
413 1. The -F script is called now *only* at the end of a tape,
414 not at the end of the dump. Thanks to Christian Haul
415 <haul@informatik.tu-darmstadt.de> for the bug report.
416
417 Normally, the device name and the current volume number
418 are passed on the command line of the script. However,
419 if you want the old style script (with no arguments
420 passed on the command line), you can enable it in
421 configure with the --enable-oldstylefscript.
422
423 2. Use posix signal handling to preserve dump functionality
424 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
425 the patch.
426
427 3. Made the exit code of restore in case of a 'C'ompare
428 command reflect the result of the compare. An exit status
429 of 0 means the dump archive is correct, 1 means tape errors,
430 2 means that some files were modified. Thanks to Kenneth Porter
431 <shiva@well.com> for the suggestion.
432
433 4. Made (finally) quotes work in interactive restore.
434
435 5. Small fixes in order to allow dump to compile with a
436 really old e2fsprogs version. Thanks to Ian Zimmerman
437 <itz@speakeasy.org> for the bug report.
438
439 6. Add GNU readline capabilities to interactive restore.
440 Use configure's --enable-readline flag to enable this feature.
441 Thanks to Patrik Schindler <poc@pocnet.net> for the
442 suggestion.
443
444 7. Do the compare on the fly in restore 'C' mode (this will
445 allow not to exhaust the available /tmp space when
446 ccmparing large files). Thanks to Kenneth Porter
447 <shiva@well.com> for the suggestion.
448
449 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
450 ====================================================================
451
452 1. Fixed some several duplicate 'DUMP: DUMP:' in the
453 output of dump.
454
455 2. Corrected the estimation of blocks to dump. Note that
456 this is still wrong for large directory inodes, due
457 to the size of a BSD directory entry on the tape
458 (estimation is based on the size of the ext2 directory,
459 which is smaller because it doesn't contain the null
460 character at the end).
461
462 3. Made dump report the total number of MB written to
463 tape at the end of run. Thanks to W. Reilly Cooley
464 <wcooley@nakedape.cc> for the patch.
465
466 4. Added the -X option to restore giving the possibility
467 to read the names of the files to be extracted/listed
468 from a text file (in addition of the command line).
469 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
470 patch.
471
472 5. Added the device name and the current volume number
473 as arguments to the end of tape script (dump -F option).
474
475 6. Made the multi-volume dump work again (after having
476 broken it in 0.4b15).
477
478 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
479 ===================================================================
480
481 1. Added a prompt command in interactive restore mode. Thanks
482 to Andreas Dilger <adilger@home.com> for the patch.
483
484 2. Fixed a buffer overflow problem in dump (caused by
485 not checking the size of the filesystem parameter).
486 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
487 reporting this on Bugtraq (and to several dump users
488 who forwarded me his mail).
489
490 3. Added the '-F script' option to dump in order to
491 launch a script at the end of each tape (to be used
492 with a tape changer, or to notify the sysadmin by
493 pager etc.).
494
495 4. Fixed a bug in restore compare code caused by the changes
496 I made in 0.4b14.
497
498 5. Fixed the treatment of options using the old BSD syntax
499 in both dump and restore.
500
501 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
502 =======================================================================
503
504 1. Fixed a bug in dump which may have caused invalid deleted
505 directories to be dumped out if they were deleted after the
506 mapping pass. This could occure on active filesystem and lead
507 to heap corruption (causing dump malfunction in many possible ways).
508 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
509 bug and submitting the patch.
510
511 2. Fixed the handling of the filesystem parameter in dump. A
512 '/mnt/disk' parameter caused the disk contents to be dumped,
513 but a '/mnt/disk/' parameter caused the mountpoint directory
514 to be dumped (generally an empty directory).
515
516 3. Improved the output of dump in order to tell which directory
517 it is currently dumping (when dumping a subtree).
518
519 4. Added the '-e' exclude inode option to dump. Thanks to
520 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
521
522 5. Added a REPORTING-BUGS file in order to provide a guide
523 on how to correctly report a bug in dump/restore.
524
525 6. Fixed a restore problem when restoring a hard link to an inode
526 having the immutable or append-only attribute set. Thanks to
527 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
528
529 7. Fixed a compatibility problem between dumps made on little
530 endian machines (the format was correct) and big endian
531 machines (the format was incorrect). This fix break the
532 compatibility with the older dumps made on big endian
533 machines (sparc, m86k, ppc etc). For the first time in
534 linux dump's history (I believe), the dumps made by *BSD,
535 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
536 so interchangeable. Thanks to Rob Cermak
537 <cermak@ahab.rutgers.edu> for submitting the bug and
538 helping me test the fix.
539
540 8. Fixed the way dump reports the remaining percent/time, if
541 the number of blocks actually dumped exceeds the estimated
542 number of blocks. Thanks to Jean-Paul van der Jagt
543 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
544
545 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
546 ======================================================================
547
548 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
549 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
550
551 2. Removed the requirement to build the RPM as root from the
552 spec file. Thanks to Christian Weisgerber
553 <naddy@mips.rhein-neckar.de> for submitting this.
554
555 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
556 causing dump to block when dumping really huge filesystems.
557 Many thanks to Patrik Schindler <poc@pocnet.net> for
558 helping me find this bug.
559
560 4. Fixed the treatment for an interrupt signal when dump access
561 the remote tape through RSH. Thanks to Christian Weisgerber
562 <naddy@mips.rhein-neckar.de> for providing the patch.
563
564 5. Fixed a bug which was causing dump/restore to display
565 garbage characters instead of the remote host name.
566
567 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
568 =====================================================================
569
570 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
571 <kukuk@suse.de> for submitting the patch.
572
573 2. Fix for the exit code when using the size estimate option of
574 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
575 submitting the patch.
576
577 3. Handle EINTR in atomical reads/writes in dump, which was causing
578 dump to fail on some systems. Thanks to Eric Jergensen
579 <eric@dvns.com> for reporting the bug and submitting the patch.
580
581 4. Handle more than 16 characters for the device names in dumpdates.
582 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
583 tracking down the problem and proposing the solution.
584
585 5. Fixed a bug in dump which prevented the creation of the
586 dumpdates file when doing a 0-level dump without already
587 having a dumpdates file. Thanks to Patrik Schindler
588 <poc@pocnet.net> for reporting the bug.
589
590 6. Changed the way dump 'S' flag reports the size estimate
591 from number of blocks into bytes (making it compatible
592 with the Solaris version, and simplifying things for
593 amanda users). Thanks to Jason L Tibbitts III
594 <tibbs@math.uh.edu> for reporting the bug.
595
596 7. Fixed a compatibility problem in linux/alpha dump tape format.
597 Now the linux/alpha dump are (again) compatible with the
598 other dump formats. But this breaks compatibility with
599 older dumps made on alpha. Thanks to Mike Tibor
600 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
601
602 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
603 ======================================================================
604
605 1. Added a '--enable-kerberos' to configure.
606
607 2. Added a 'S' option to dump which determines the amount of space
608 that is needed to perform the dump without actually doing it, similar
609 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
610 <cermak@ahab.rutgers.edu>.
611
612 3. Added a 'M' multi-volume option to dump and restore which enables
613 dumping to multiple files (useful when dumping to an ext2
614 partition to make several dump files in order to bypass the 2GB
615 file size limitation). The 'f' argument is treated as a prefix and
616 the output files will be named <prefix>001, <prefix>002 etc. With
617 the 'M' flag, restore automatically selects the right file without
618 asking to enter a new tape each time.
619
620 4. Fixed a memory leak which was causing dump to grow very big
621 (270MB when dumping a 10GB filesystem...). Thanks to Jason
622 Fearon <jasonf@netrider.org.au> for reporting the bug.
623
624 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
625 ======================================================================
626
627 1. Make configure test if the system glob routines support
628 extended syntax (ALTDIRFUNC). If not, use the internal glob
629 routines instead of system ones. Thanks to Bernhard Sadlowski
630 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
631 and helping me resolve this and other minor libc5 compiling
632 glitches.
633
634 2. Fix a problem when dumping a ext2fs with the 'filetype'
635 feature enabled. Thanks to Patrick J. LoPresti
636 <patl@cag.lcs.mit.edu> for reporting the bug and to
637 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
638
639 3. Made the nodump flag work on directories. A directory which
640 has the nodump flag gets never dumped, regardless of its
641 contents.
642
643 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
644 which allows dump on an active ext3 filesystem. However, this
645 is a "quick and dirty" patch which enables backup of an ext3
646 filesystem through the ext2 compatibility (by ignoring the
647 NEEDS_RECOVERY bit). The journal file is not recognized and
648 it is dumped (it should not).
649
650 5. Test the superblock compatibility flags when dumping, in order
651 to be sure that we know how to deal with specific features.
652
653 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
654 ====================================================================
655
656 1. Use lchown instead of chown, fixing a possible security problem
657 when restoring symlinks (a malicious user could use this
658 to deliberately corrupt the ownership of important system files).
659 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
660 this and providing the patch.
661
662 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
663 ====================================================================
664
665 1. Put dump sources under CVS, added Id tags in all files so
666 one can use 'ident' on binary files.
667
668 2. Added the dump/restore version in the usage text so one can
669 easily verify the version he is using.
670
671 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
672 a va_start/va_end problem on linux-ppc (always call va_start
673 va_end in pairs each time we use a vararg function).
674
675 4. Added again the DT_* constants because old libc does not
676 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
677 for submitting the bug report.
678
679 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
680 this should enable dumping big (huge) filesystems.
681
682 6. Added the RSH environment variable in order to be able to
683 use a rsh replacement like ssh when doing remote backups (and
684 bypass the security limitations of rcmd). Now you can do remote
685 backups without being root (or making dump setuid root).
686
687 7. Modified again the way dumpdates works. For incremental dumps,
688 we need to read dumpdates even if we are not using 'u' option.
689 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
690 this should work.
691
692 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
693 ===================================================================
694
695 1. Removed the 'k' flag from the restore 'about' text if kerberos
696 was not compiled in.
697
698 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
699 to them (fsetflags takes a char*, setflags an open fd!).
700
701 3. (f)setflags is called only if the flags aren't empty. If the
702 file is a special file, a warning is printed, because changing
703 flags implies opening the device. Normally, a special file
704 should not have any flag... (Debian bug #29775, patch provided
705 by Abhijit Dasgupta <abhijit@ans.net>).
706
707 4. Made possible to dump a file system not mentioned in /etc/fstab.
708 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
709
710 5. Changed the default behaviour to not create dumpdates
711 unless 'u' option is specified. Removed the old "debian-patch"
712 which provided the same thing. (Debian bug #38136, #33818).
713
714 6. Removed all those dump*announce, since they were getting old...
715
716 7. Added warning messages if dumpdates does not exist and
717 when an inferior level dump does not exist (except for a level 0
718 dump).
719
720 8. Debugged the glob calls in interactive mode: restore used a
721 dirent struct which was different from the /usr/include/dirent.h
722 one (this used to work, is it a glibc2 change?), so none of the
723 compat glob (which used /usr/include/dirent.h) or the system glob
724 worked. Restore use now the system dirent (and the system
725 DT_* constants), which are compatible with BSD ones.
726
727 9. Added a configure flag (--with-dumpdatespath) to specify
728 the location of dumpdates file. By default, it is
729 /etc/dumpdates.
730
731 10. Added the "AUTHOR" and "AVAILABILITY" sections and
732 included the current date/version in man pages.
733
734 11. Corrected the estimation of remaining time when
735 the operator doesn't change the tapes quickly enough. This
736 was an old bug, I thought I corrected it, and discovered
737 that in fact it was corrected in two different places, so
738 the results canceled each other...
739
740 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
741 ===================================================================
742
743 1. Integrated multiple patches from RedHat, Debian and SuSE:
744
745 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
746 isn't specified.
747 - several fixes in the man pages.
748 - update the default tape device to /dev/st0.
749 - many updates for Linux Alpha (byte ordering, size_t etc).
750 - buffer overruns.
751 - use environment variable for TMPDIR (instead of /tmp).
752 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
753 - workaround egcs bug (RedHat bugs #4281 and #2989).
754 - wire $(OPT) throughout Makefile's.
755
756 2. Upgrade the dump revision to 1, making possible to dump filesystems
757 made with e2fsprogs-1.15 or newer. Nothing seems to break...
758
759 3. Fix some compile warnings, prototype all functions.
760
761 4. Use glibc err/glob instead of internal compatibility
762 routines (only if available).
763
764 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
765 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
766
767 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
768 ======================================================================
769
770 1. Integrated the changes from FreeBSD-3.1-RELEASE
771 (mostly bug fixes, buffer overruns, dump has now an "automatic
772 tape length calculation" flag, dump/restore can use kerberos now
773 (this is NOT tested), use environment variables for TAPE and
774 RMT etc.).
775
776 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
777
778 3. Corrected a bug in restore when using 'C' option with multi-volumes
779 tapes (files splited accros two tapes give "size changed" errors
780 when comparing).
781
782 4. Corrected the long standing bug when dumping multiple tapes.
783 This works for me, needs further testing.
784
785 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
786 ====================================================================
787
788 1. Dump now runs correctly on kernels 2.1.x
789 Fix made by Gerald Peters <gapeters@worldnet.att.net>
790
791 Changes between versions 0.4b2 and 0.4b3
792 ========================================
793
794 1. Use realpath() if available
795
796 2. Report statistics
797
798 Changes between versions 0.4b1 and 0.4b2
799 ========================================
800
801 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
802 the patch)
803
804 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
805
806 3. Fixed configure.in to correctly handle --enable-rmt
807
808 Changes between versions 0.3 and 0.4b1
809 ======================================
810
811 1. Integrated the changes from 4.4BSD-Lite2
812
813 2. Integrated the patches from Debian and Red Hat
814
815 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
816
817 4. Changed dump to use the Ext2fs library to get block addresses. This
818 should solve the endianness problem on SparcLinux.
819
820 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
821 distribution's one) to use autoconf
822
823 6. Fixed a few minor bugs
824
825 Changes between versions 0.2e and 0.2f
826 ======================================
827
828 1. Added the creation of named pipes (fifos) in restore.
829
830 2. Added the -N flag in the restore manual page.
831
832 3. Added the file kernel.patch which contains the llseek() optimization
833 patch for 1.2.x kernels.
834
835 4. Fixed a bug in the restoration of symbolic links: owner and group were
836 not restored.
837
838 5. Integrated some changes from FreeBSD 2.2.
839
840 6. Added a call to ftruncate() after restoring each file to restore
841 correctly files ending by a hole.
842
843 Changes between versions 0.2d and 0.2e
844 ======================================
845
846 1. Fixed a bug in the "set owner/mode" process. Every file was restored
847 with owner = root (0) and group = root/wheel/whatever (0).
848
849 Changes between versions 0.2c and 0.2d
850 ======================================
851
852 1. Dump is now able to backup 2GB+ filesystems.
853
854 2. Dump and restore can now be linked as static binaries.
855
856 Changes between versions 0.2b and 0.2c
857 ======================================
858
859 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
860
861 Changes between versions 0.2a and 0.2b
862 ======================================
863
864 1. Really fixed the bug that I should have corrected in 0.2a.
865
866 2. Enabled optimization again.
867
868 Changes between versions 0.2 and 0.2a
869 =====================================
870
871 1. Disabled the optimization during compilation.
872
873 Changes between versions 0.1 and 0.2
874 ====================================
875
876 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
877 the fs_type field (actually, I modified the file fstab.c to make it
878 use the mntent functions).
879
880 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
881 made by dump should be readable by the BSD restore and backups made
882 by the BSD dump should be readable by restore. Unfortunately, this
883 means that the dump format has changed between version 0.1 and version
884 0.2 :-(
885
886 3. Dump is now able to backup a subtree, it is no longer limited to whole
887 filesystems like the BSD version.
888
889 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
890 than 2 GB.
891
892 Changes between versions 0.0 and 0.1
893 ====================================
894
895 1. Now create links rdump and rrestore during the `make install' step.
896
897 2. Linux port specific bugs added to the manual pages
898
899 3. Incorrect estimation of the number of tapes blocks fixed when doing
900 incremental backups.
901
902 4. Better ls-like format in restore in interactive mode.