[dump.git] / CHANGES
1 $Id: CHANGES,v 1.308 2010/03/08 10:57:24 stelian Exp $
3 Changes between versions 0.4b42 and 0.4b43 (released ?????????????)
4 ===================================================================
6 1. Fix a bug in dump making impossible to handle large toc files
7 (> 2 GB). Thanks to X DUGi <xdugi@users.sourceforge.net> for
8 reporting the bug (Sourceforge bug #2820629)
10 2. Fix 'restore -x' in multi-volume mode, which caused files being
11 spanned on the first and second volumes to be incorrectly extracted.
12 Thanks to Adam Tkac <atkac@redhat.com> for reporting the bug.
13 (there may be several bug reports about this issue, including
14 Sourceforge bug #2117008, RedHat bugs #232415 and #444958).
16 3. Remove -lselinux -lsepol from the standard libraries list,
17 they shouldn't be needed if selinux is not enabled in dump
18 (this also fixes a build failure when selinux libraries are
19 not installed).
21 4. Fix restoration of extended attributes of fifos and device nodes.
22 Thanks to Igor Zhbanov <izh1979@gmail.com> for the bug report.
24 5. Don't attempt to set extended attributes on files that were not
25 extracted in 'restore -N' mode. Thanks to Jan Görig
26 <jgorig@users.sourceforge.net> for the patch.
28 6. Make the dump man page explicit about ext4 support.
30 Changes between versions 0.4b41 and 0.4b42 (released June 18, 2009)
31 ===================================================================
33 1. Fix printout of the 'Connection to' message in dump/restore.
34 Thanks to Dale Wiles <dwiles@users.sourceforge.net> for the
35 bug report and its fix.
37 2. Fix dump -w/-W output for filesystems having the last backup
38 level equal to 0. Thanks to Pascal Bouchareine
39 <kalou@users.sourceforge.net> for reporting the bug.
41 3. Enable restore to handle restoring onto a different SELinux
42 policy from the dump. Thanks to Tony Nelson
43 <tony_nelson@users.sourceforge.net> for the patch.
44 (see http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=189845
45 for details)
47 4. Fix EA set failures when restoring immutable files. Thanks
48 to Andrew Kroeger for the patch.
50 5. Fix "mode file too large" errors when restoring huge backups.
51 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
52 report.
54 6. Add the ability to link against libtinfo library if present,
55 instead of libtermcap or libncurses libraries. Thanks to
56 Adam Tkac <atkac@redhat.com> for the patch.
58 7. When comparing a backup, do not consider to be an error the fact
59 that we encounter files created while dump was in progress.
60 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
61 report.
63 8. Force '-a' to be the default value when running 'restore -C'.
64 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
65 report.
67 9. Many improvements to the 'cron_dump_to_disk' example.
68 Thanks to Aaron S. Hawley <ashawley@users.sourceforge.net> for
69 the patch.
71 10. Fix the QFA generation when extended attributes are backuped
72 along with the inodes. Prior to this fix, the entries for
73 some inodes may get corrupted (duplicated entries or incorrect
74 ones), making the QFA file unusable for repositionning in
75 restore. Many thanks to Kenneth Porter <shiva@sewingwitch.com>
76 for the bug report and the testing of the fix.
78 11. Fix EA restoration in debug mode (Sourceforge bug #1986216).
80 12. Stefan Auracher <star7@onlinehome.de> noticed that the Tower
81 of Hanoi backup strategy documented in the man page wasn't
82 actually based on the Tower of Hanoi algorithm. The man page
83 was updated accordingly.
85 13. Fix missing level information from dump output when using
86 the default level (RedHat bug #493635). Thanks to Adam Tkac
87 <atkac@redhat.com> for the patch.
89 14. Fix the dump man page to reflect the fact that the default
90 dump level is 0 not 9 (RedHat bug #356121). Thanks to Adam Tkac
91 <atkac@redhat.com> for the patch.
93 15. Fixed a few spelling errors in the man pages (RedHat bug #489853).
94 Thanks to Adam Tkac <atkac@redhat.com> for the patch.
96 16. Fixed restore man page and program usage which incorrectly stated
97 that -P and -A were both allowed in the same invocation (RedHat
98 bug #490627). Thanks to Adam Tkac <atkac@redhat.com> for the patch.
100 17. Use sys_clone under Linux to share I/O contexts between dump
101 processes, thus drastically increasing the performance of dump
102 under CFQ (which is the default I/O scheduler used in a number of
103 distributions). Thanks to Jeff Moyer <jmoyer@redhat.com> for the
104 patch.
106 18. Add (preliminary) ext4 support - thanks to libext2fs which does
107 all the job for us. Thanks to Gertjan van Wingerde
108 <gwingerde@gmail.com> for the patch.
110 Changes between versions 0.4b40 and 0.4b41 (released January 2, 2006)
111 =====================================================================
113 1. Fix restore of dumped Access Control Lists. The previous
114 code for EA works fine, but ACLs needed conversion from the
115 ext2/3 disk format to posix_acl format before restoring.
117 2. Fix some issues with restoration of EA on big endian
118 platforms.
120 3. Fix restore when the symtab is over 2GB in size.
122 4. Made the directory hash indexing an optional feature,
123 accessible by the '-H' option of restore, and disable it
124 by default.
126 5. Fixed dump to not include extended attributes information
127 in the toc (archive) file which confused restore -t.
129 Changes between versions 0.4b39 and 0.4b40 (released May 2, 2005)
130 =================================================================
132 1. Changed restore to emit warnings (instead of emitting a fatal
133 error) if a file (or a directory) is unavailable for a
134 comparision (if the user doesn't have the necessary permissions
135 to access it for example). Thanks to Kenneth Porter
136 <shiva@sewingwitch.com> for the bug report.
138 2. Re-done the 'do not save directory entries to non-dumped inodes
139 (excluded from dump)' feature. The previous implementation
140 worked well for excluded directories but not for regular files.
141 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
142 report.
144 3. Fixed a bug in dump where the tape size was miscalculated when
145 the user used -d/-s to specify the tape characteristics. Thanks
146 to Philip Goisman <goisman@physics.arizona.edu> for reporting
147 the bug.
149 4. Fixed another bug introduced in restore with the hashtree
150 implementation. This one caused restore to stop saying
151 "removenode: non-empty directory" in some cases.
153 5. Added support for dumping and restoring ext2/3 extended
154 attributes (EA), like the access control lists (ACL) or
155 the security labels used by SELinux.
157 Changes between versions 0.4b38 and 0.4b39 (released January 21, 2005)
158 ======================================================================
160 1. The newly added dump_on_cd_3 example was buggy, replace it
161 with an updated version from Andrew Basterfield
162 <bob@cemetery.homeunix.org>.
164 2. Made restore to chdir() back into the initial directory when
165 dumping core while aborting a comparision operation. The
166 previous behaviour was to write the corefile at the root of
167 the directory being compared, which could very well be
168 read only and preventing the corefile generation. Thanks
169 to Kenneth Porter <shiva@sewingwitch.com> for the bug report.
171 3. Silenced the failure to call fgetflags() when comparing an
172 entry which has no ext2 attributes (as in lsattr()).
174 4. Fix a brown paper bug in restore -C which broke restore and
175 caused modifications on the filesystem being compared
176 (directories containing a file with the same name as the
177 directory get renamed to RSTTMP...). Thanks to Kenneth Porter
178 <shiva@sewingwitch.com> for finding the bug and helping me
179 reproduce it.
181 5. Made restore -C force the -N flag (no writing allowed on
182 the disk) in order to prevent more bugs like the above one.
184 Changes between versions 0.4b37 and 0.4b38 (released January 7, 2005)
185 =====================================================================
187 1. Fix a couple of troff syntax bugs in the man pages.
188 Thanks to Eric Raymond <esr@thyrsus.com> for the patch.
190 2. Made restore use either libncurses or libtermcap, depending
191 on which one is available at configure time.
193 3. Fixed restore negative size display bug when comparing a
194 dump containing files over 2GB. Thanks to Steve Bonds
195 <sbonds@users.sourceforge.net> for the bug report.
197 4. Do not save directory entries to non-dumped inodes
198 (excluded from dump). This will eliminate the 'missing
199 file' warnings when doing 'restore -C'.
201 5. Fix dump crash when backuping a huge (2TB) filesystem,
202 due to a bogus calculation on the inode map size.
203 Thanks to Kevin B. Haines <K.B.Haines@rl.ac.uk> for
204 submitting the bug and testing the fix.
206 6. Fix a problem in restore where the final \0 in the symbolic
207 link names could have been lost, generating corrupt filenames.
208 Thanks to Kyle Wilson <kyle.wilson@amd.com> for reporting the
209 bug.
211 7. Implemented a hash list for the directory names in restore.
212 The linear list used before caused problems in interactive
213 restores when dealing with directories having thousands of
214 entries. Thanks to Brian Ristuccia <bristuccia@starentnetworks.com>
215 for reporting the bug.
217 8. Improved restore -C, this time including the directory
218 attributes into the comparision.
220 9. Made restore understand tapes containing EA/ACLs (which will
221 be dumped by the next version of dump). In this version
222 extended attributes on the tape are ignored, for full EA/ACL
223 support wait for the next version or try the experimental EA
224 patch.
226 Changes between versions 0.4b36 and 0.4b37 (released July 7, 2004)
227 ==================================================================
229 1. Added the --enable-staticz configure option which enables
230 dump and restore to be linked with static versions of
231 libz and libbz2 (and dynamic versions of all other libraries).
232 This will make Debian users happy, because libz and libbz2
233 were the only needed libraries living in /usr, all the
234 others live in /lib. In case of system emergency, it is
235 better not to have to rely on an extra filesystem.
236 Thanks to Bdale Garbee <bdale@gag.com> for the suggestion.
238 2. Fix compilation on (at least the Linux Debian port to) AMD64.
239 (<ext2fs/ext2_types.h> defines some types (__s64 and __u64)
240 that are also defined by <linux/types.h> (<asm/types.h>) and
241 they conflict).
243 3. Make dump's reading of the dumpdates file a bit more robust,
244 preventing dump from crashing when the dumpdates file has
245 been modified by hand.
247 4. Fixed some offset calculations in dump code which could
248 lead to "bread lseek errors" on large filesystems. Thanks
249 to Bruce Lowekamp <lowekamp@users.sourceforge.net> for
250 reporting this bug and debugging the issue.
252 5. Made dump use the blkid library when searching for devices
253 by label or uuid instead of dump's own routines.
255 6. Corrected a bug in dump where a wrong LABEL=... line in
256 /etc/fstab could prevent dump from dumping unrelated
257 filesystems. Thanks to Bruce Lowekamp
258 <lowekamp@users.sourceforge.net> for reporting the bug.
260 Changes between versions 0.4b35 and 0.4b36 (released April 21, 2004)
261 ====================================================================
263 1. Fixed dump compilation with old gcc versions. Thanks to
264 Mike Castle <dalgoda@users.sourceforge.net> for the patch.
266 2. Fixed some warnings (howmany, roundup, powerof2 redefined)
267 when compiling against a recent glibc version.
269 3. Fixed a bug in restore preventing the read of a dump tape
270 written with Solaris 7 ufsdump. Thanks to Patrick Higgins
271 <phiggins@transzap.com> for reporting the bug and providing
272 the test case.
274 4. Changed dump to enable the creation of volumes of different
275 sizes in a single run (make -B accept a list of values).
276 Patch contributed by Florian Zumbiehl <florz@gmx.de>.
278 5. Use the glibc provided minor() and major() macros instead
279 of our own bitmask implementation. This should be safe for
280 when the major/minor namespace will migrate to 32 bits.
281 Thanks to Zhang Jun <zhangjun@nanjing-fnst.com> for reporting
282 the bug.
284 6. Made explicit in the dump man page that dump will not create
285 a remote file, it will only write to an already existing one.
287 7. Another try at making size estimates better again.
289 8. Put back the inconditional running of the end-of-tape
290 script which was changed in 0.434 to be run only when -M
291 or multiple -f were NOT specified. Some users rely on this
292 feature even when it is combined with -M/-f.
294 9. Fixed restore when restoring huge backups (where rstdir...
295 temporary files are over 2GB). Thanks to
296 Raphael Jaffey <rjaffey@artic.edu> for reporting this,
297 debugging the issue and making the patch.
299 10. Made restore understand FreeBSD UFS2 tapes. Thanks to
300 David <vrtin@users.sourceforge.net> for submitting the bug
301 and providing a test case.
303 11. Made dump work with an arbitrary integer level (no more 10
304 levels only limitation). Thanks to Matthew
305 <msvincen@midway.uchicago.edu> for the patch.
307 Changes between versions 0.4b34 and 0.4b35 (released December 21, 2003)
308 =======================================================================
310 1. Added a note in the dump man page saying that the default
311 blocksize can be 32 if -d is used with a high density argument.
312 Thanks to Antonios Christofides <A.Christofides@itia.ntua.gr>
313 for the patch.
315 2. Fixed configure to correctly understand CPPFLAGS, CFLAGS,
316 LDFLAGS environment variables. Thanks to Arcady Genkin
317 <antipode@users.sourceforge.net> for reporting the bug.
319 3. Made -e/-E options of dump accept an unlimited number of inodes
320 to be excluded, rather than a hardcoded maximum. Thanks to
321 Dietrich Rothe <d-rothe@users.sourceforge.net> for the patch.
323 4. Updated the autoconf system to 2.50. Forced the -D_BSD_SOURCE
324 and -D_USE_BSD_SIGNAL defines in configure in order to solve
325 64bit build problems because quad_t is redifined with a
326 different signature. Thanks to Mike Harris <mharris@redhat.com>
327 for reporting this bug.
329 5. Made restore build on Solaris, making possible to
330 restore Linux's "enhanced" tapes. Thanks to Uwe Gohlke
331 <uwe@ugsoft.de> for the patch.
333 6. Made an extension in the dump tape format capable of saving
334 MacOSX specific inode extensions. Uwe Gohlke <uwe@ugsoft.de>
335 wrote the extension and contributed the restore code back
336 into this codebase. The same extension mechanism will be
337 used in the future to save ACLs...
339 7. Made rmt work correctly with regard to QFA and local/remote
340 files and tapes. The remote access will however work only
341 when the dump provided rmt version is used. If you want to
342 use another rmt server, please do not use the QFA feature.
343 Thanks to Uwe Gohlke <uwe@ugsoft.de> for the patch.
345 Changes between versions 0.4b33 and 0.4b34 (released April 18, 2003)
346 ====================================================================
348 1. Fixed the INSTALL file to reflect the actual install paths.
349 Thanks to David Ronis <ronis@ronispc.chem.mcgill.ca> for
350 reporting the bug.
352 2. Fixed the configure script to only check for headers presence
353 instead of trying to compile them. This should fix issues
354 with old build environments. Thanks to Kari Lempiainen
355 <kari@funky.pp.fi> for reporting the bug.
357 3. Fixed restore to correctly ignore sockets when comparing
358 a dump (as socket cannot be properly restored anyway). Thanks
359 to Gunther Reiszig <gunther@mit.edu> for reporting the bug.
361 4. Fixed restore to correctly access the archive file (-A argument)
362 even when using a remote tape. Thanks to Fabrice Bellet
363 <fabrice@bellet.info> for reporting the bug.
365 5. Fixed (again) handling of long (largefile) seeks in rmt.
366 Thanks to Fabrice Bellet <fabrice@bellet.info> for reporting
367 the bug.
369 6. Fixed restore corner case when dealing with large block sizes
370 dump is able to write now (-b 1024). Thanks to Fabrice Bellet
371 <fabrice@bellet.info> for reporting the bug.
373 7. Fixed a bug preventing dump to access a filesystem having
374 a label exactly 16 bytes in length. Thanks to <minduni@ti-edu.ch>
375 for reporting the bug.
377 8. Made dump store dump dates using explicit timezones, fixing a
378 problem with incremental dumps if the timezone is modified
379 between the dumps. Thanks to Philippe Troin <phil@fifi.org> for
380 the bug report and the patch.
382 9. Fixed a bug encountered when dumping individual files (not full
383 filesystems or directories) and dangling symbolic links happen
384 to be in the list of files. For as far as dump is concerned,
385 dangling symbolic links are allowed, and are dumped as is.
386 Thanks to Jin-su Ahn <jsahn@ee.snu.ac.kr> for reporting the
387 bug and providing the fix.
389 10. Fixed open and creation modes and permissions for QFA and
390 table-of-contents files in dump and restore. Thanks to
391 Philippe Troin <phil@fifi.org> for the patch.
393 11. Fixed the archive file descriptor handling enabling it to be 0.
394 This can happen in some cases when shell redirections are used.
395 Thanks to Philippe Troin <phil@fifi.org> for the patch.
397 12. Delayed the opening of archive file until after suid had been
398 dropped (fixing a possible security issue if dump is suid).
399 Thanks to Philippe Troin <phil@fifi.org> for the patch.
401 13. Fixed the 'S' command handling in the rmt client part.
402 Thanks to Philippe Troin <phil@fifi.org> for the patch.
404 14. Modified the end-of-tape script handling to print out statistics
405 (and stop the timer) before launching the eot script. Also, the eot
406 script does not get run anymore when using -M (which makes sense) or
407 when multiple tapes are listed on the command line
408 (-f tape0,tape1,tapen) (which also makes sense).
409 Thanks to Philippe Troin <phil@fifi.org> for the patch.
411 15. Relicensed dump/restore under the 'revised' BSD license, as per
412 ftp://ftp.cs.berkeley.edu/ucb/4bsd/README.Impt.License.Change.
414 16. Added LZO compression to dump. This new compression method has
415 the advantage of being super fast, thus not killing tape streaming
416 on slow machines. Thanks to Helmut Jarausch
417 <jarausch@igpm.rwth-aachen.de> for the patch and to
418 Markus Oberhumer <markus@oberhumer.com> for giving special permission
419 to include his miniLZO project (GPL licensed) in dump/restore.
421 17. Some small buffer overruns fixes in rmt. Thanks to Antonomasia
422 <ant@notatla.demon.co.uk> for reporting the bugs.
424 18. Added a special rmt version which can do encryption when writing
425 to tape. Read examples/encrypted_rmt/README for details on
426 how to enable and configure it. Thanks to Ken Lalonde
427 <ken@globalremit.com> for the patch.
429 19. Made dump work with 2.5 kernel end of tape early warning semantics.
430 Thanks to Kai Makisara <Kai.Makisara@kolumbus.fi> for the patch.
432 20. Fixed a bug which caused dump -w|-W not to work anymore, because
433 the fs_freq and fs_passno fields in /etc/mtab are always set
434 to 0 0. Thanks to Trent Piepho <xyzzy@speakeasy.org> for
435 reporting the bug.
437 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
438 =======================================================================
440 1. Added a note in the restore man page clarifying the question
441 'set the permissions on the current directory ?' asked by
442 restore at the end of treatment in -i and -x modes.
444 2. Fixed the endianess issues when reading compressed tapes.
445 Thanks to Dark Force <daq4th@users.sourceforge.net> for
446 reporting this bug and providing test cases.
448 3. Fixed the "ACL won't be dumped" warning message (which showed
449 an extra, unrelated error message). Thanks to Dragan Krnic
450 <dkrnic@lycos.com> for reporting this bug.
452 4. Made dump look first into /etc/mtab, then into /etc/fstab
453 when searching for filesystem paths. Also fixed some problems
454 caused by binding mounts or multiple block device mounts.
455 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
456 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
457 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
458 and #160305, Sourceforge bugs #618699 and #679832).
460 5. Made dump's -I option accept the value '0' meaning all the
461 read errors will be ignored. This can be useful when running
462 dump from unattended sessions (like cron jobs). Thanks to
463 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
465 6. Fixed the output of dump to indicate 'blocks' instead of
466 'tape blocks' in the various messages (blocks are always
467 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
468 made it clearly print the current blocksize at the start of
469 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
470 the suggestions.
472 7. Made rmt understand long (largefiles) seeks.
474 8. Fixed build with very old versions of libext2fs, where
475 EXT2_FT_* constants were undefined.
477 9. Made dump accept the dumpdates path on the command line
478 (-D file option) instead of using only the hardcoded one.
479 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
480 suggestion.
482 10. Enabled rmt, LFS, readline, QFA options by default in
483 ./configure. Updated the configure process (new versions
484 of config.guess, config.sub etc).
486 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
487 =======================================================================
489 1. Changed dump to use fcntl(F_SETLK) style locking instead
490 of flock() when locking the dumpdates file. With the old
491 locking scheme, a local user having read rights on the
492 dumpdates file could be able to do a Denial of Service attack
493 on dump. In order to lock the dumpdates file with the new
494 scheme, the user would need to have write access on the file.
495 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
496 reporting the bug (originally a bugtraq post).
498 2. Fixed interactive 'ls' which caused spurious errors warnings
499 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
500 <jorgen_ostling@users.sourceforge.net> for reporting this
501 bug.
503 3. Fixed dump's estimate when dealing with sparse inodes.
505 4. Modified dump to allow setting a blocksize bigger than 32kB
506 (raised the limit to 1024kB), because newer hardware needs
507 this for proper operation. Thanks to Dirk Traenapp
508 <dtraenapp@users.sourceforge.net> for reporting this.
510 5. Fixed a bug causing Dump to stop and report an error if an
511 inode number in the exclude file was followed by some amount
512 of whitespace. Thanks to Jeffrey Sofferin
513 <sofferin@users.sourceforge.net> for reporting this bug.
515 6. Fixed a bug which caused restore, in some particular cases,
516 to ask some 'scary' questions and leave a bunch of RSTTMP
517 directories behind when restoring incremental tapes. Thanks
518 to Philippe Troin <phil@fifi.org> for reporting this bug and
519 providing the test cases.
521 7. Changed the wording when inodes are excluded from dump:
522 replaced 'Added inode 7 to exclude list' with
523 'Excluding inode 7 (resize inode) from dump', as suggested
524 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
526 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
527 ===================================================================
529 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
530 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
531 <eros@lamel.bo.cnr.it> for reporting the bug.
533 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
534 ===================================================================
536 1. Made dump print out the ext2 logical block number in case of a read
537 error, which can be given as an argument to debugfs' ncheck command,
538 in order to find out the affected inode. Added note about this
539 usage in dump's man page.
541 2. Fixed a problem in restore when reading tapes written on big
542 endian machines with very old versions of dump. The patch was
543 contributed by George Helffrich <george@geology.bristol.ac.uk>.
545 3. Fixed the tape length calculation when using large tapes
546 and compression. Thanks to Georg Lippold
547 <g_lippold@sourceforge.net> for reporting the bug.
549 4. Added a new set of examples in dump_on_cd_2 directory, based
550 on dump_on_cd examples but somewhat enhanced, supporting DVD
551 media, and localized in english and german. Thanks to
552 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
554 5. Made dump save 32 bit UID/GID. Internally, this was achieved
555 by switching from the old BSD inode format to the new BSD
556 inode format, which means that the tape format was changed.
557 However, since all restore versions out there should
558 transparently support both inode formats, the change should
559 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
560 reporting the bug.
562 6. Fixed a lot of warnings in the code shown when compiling
563 with 'gcc -W'. Thanks to Matthias Andree
564 <matthias.andree@stud.uni-dortmund.de> for reporting this.
566 7. Fixed a small markup bug in the dump man page. Thanks to
567 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
568 patch.
570 8. Rewrote entirely the man pages using the tmac.an macro
571 package (Linux man page format) instead of the original BSD
572 format. They should be now cleaner and easier to modify.
574 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
575 ==================================================================
577 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
578 non Linux clients were misinterpreted. The description of the problem
579 (incompatible numbering in Linux mtio opcodes) is documented at
580 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
581 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
582 providing an excellent, cross-platform replacement for rmt in his
583 star package.
585 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
586 which caused restore to fail when the inode maps were not entirely
587 contained on the first volume (when using really small volumes or
588 when dumping a huge number of inodes).
590 3. Fixed a problem in dump, where files in subdirectories of directories
591 that have the nodump flag set (or being excluded with -e/-E) were
592 sometimes incorrectly dumped (instead of being excluded). The original
593 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
594 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
595 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
596 help testing the patch.
598 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
599 which makes one able to pipe the output of dump, by the net, to
600 a remote CD-burner server.
602 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
603 and use GNU's symbolic syntax over rmt instead of numerical values
604 to assure multiple platform compatibility.
606 6. Documented the -d option in restore.
608 7. Added a -v (verbose) mode to dump. For now it just prints the number
609 of the inode being dumped, but this could evolve in future versions
610 to include interesting debugging output.
612 8. Added a -o flag to restore, which automatically restores the current
613 directory permissions in -i or -x mode, without asking the operator
614 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
615 and Peter Samuel <peters@e-smith.com>.
617 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
618 ====================================================================
620 1. Fixed a bug in the QFA feature which could have caused
621 a bad tape positionning on restore (causing extra delays in
622 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
624 2. Added a small note in the dump man page specifying that
625 there should be no white space between the option letter and
626 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
628 3. Made restore work with older versions of the readline library,
629 by checking in configure for several extended symbols. Restore
630 can now be compiled with a readline as old as the 2.0 release
631 (though it may be a good idea to upgrade it to a more recent
632 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
633 reporting the build failures.
635 4. Fixed a performance problem with the QFA file creation in
636 dump, which made unnecessary seeks on the tape slowing down
637 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
638 reporting this issue.
640 5. Removed the inclusion of some kernel headers in the dump
641 source, which prevented the compile in some kernel/glibc
642 headers/architecture combination. Thanks to Bdale Garbee
643 <bdale@gag.com> for reporting the bug.
645 6. Added the appropriate error message when dump fails to
646 open the output file for writing. Thanks to Amith Varghese
647 <amithv@yahoo.com> for reporting this bug.
649 7. Made restore able to understand large Solaris ufsdump tapes
650 (containing inodes bigger than 4194304). Sun have introduced
651 an "extension" to the dump tape format when dealing with
652 those inodes, which was uncorrectly handled by Linux restore.
653 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
654 providing a test case.
656 8. Added the -m parameter to dump which optimises the output for
657 inodes having been changed but not modified since the last dump
658 ('changed' and 'modified' have the meaning defined in stat(2)).
659 For those inodes, dump will save only the metadata, instead of
660 saving the entire inode contents. Inodes which are either
661 directories or have been modified since the last dump are saved
662 in a regular way. Uses of this flag must be consistent, meaning
663 that either every dump in an incremental dump set have the flag,
664 or no one has it.
666 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
667 =======================================================================
669 1. Fixed behaviour of dump when exceeding resource limits
670 (SIGXFSZ treatment).
672 2. Added the -L flag to restore to allow the user to specify a
673 maximal allowed number of miscompares when using restore
674 with the -C option to check the backup.
676 3. Detailed the manual entry for the -N option of restore.
678 4. Added the -a flag to restore to make able doing unattended
679 restores in -i or -x mode (automatically walks through the
680 multiple dump volumes).
682 5. Extended the QFA mode to work with local files and/or
683 remote tapes and files. This way, restore can know in advance
684 the tape number and the offset for the inodes to extract and
685 can minimize the extraction time by seeking directly to the
686 good tape/offset.
688 6. Added the -A <archive> option to both dump and restore,
689 which makes dump to archive a dump table-of-contents in
690 the specified file to be used by restore to determine
691 whether a file is in the dump file that is being restored.
692 (the archive file syntax is also compatible with the
693 Solaris ufsdump generated one).
695 7. Small fix in restore making it able to read some (broken ?)
696 Solaris ufsdump tapes.
698 8. Fixed dump to correctly recognise the root filesystem when using
699 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
700 for reporting this bug.
702 9. Added the -P <file> option to restore to create a
703 Quick File Access file from an already made dump. Patch
704 contributed by Uwe Gohlke <uwe@ugsoft.de>.
706 10. Made restore compile and run on Solaris, making it a
707 possible replacement for the standard ufsrestore. Port was
708 contributed by Uwe Gohlke <uwe@ugsoft.de>.
710 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
711 =====================================================================
713 1. Added a set of backup scripts from Eugenio Diaz
714 <getnito@yahoo.com> in the examples section. It features
715 automatic (cron based) full and incremental dumping of
716 several filesystems on a separate filesystem.
718 2. Fixed a off-by-one miscalculation which disabled dumping
719 a one letter subdirectory. Thanks to Chris Siebenmann
720 <cks@utcc.utoronto.ca> for reporting the bug.
722 3. Fixed several restore bugs occuring when trying to
723 restore some missing files on the tape. Thanks to Chris
724 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
726 4. Fixed --with-ldopts configure argument passing, installing from
727 a separate object directory, makefile cleanups contributed
728 by <splite@purdue.edu>.
730 5. Fix a bug which could caused, in some conditions, the highest
731 number inode of a filesystem, to not be dumped. Many thanks
732 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
733 chase this bug.
735 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
736 =======================================================================
738 1. Added a mini howto from Patrick Walsh in the examples
739 directory.
741 2. Minor man pages syntax corrections. Thanks to
742 Chris Danis <screechco@home.com> for reporting the bugs.
744 3. Added a script from David B. Peterson <dave@toppledwagon.com>
745 to the examples section. It features dumping several
746 filesystems to a remote tape drive upon ssh.
748 4. Added a patch provided by Richard Jones <rich@annexia.org>
749 which allows BRADEMAX (number of read errors tolerated by
750 dump) to be adjusted using the -I option.
752 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
753 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
754 patch.
756 6. Add the -l option to restore to specify if, when doing a
757 remote restore, the file used to access the backup is a
758 regular file (the defaults being a tape drive). Restore needs
759 to know this information when reading a remote compressed
760 dump. Previously, this information was autodetected, but
761 the autodetection code fails (with ioctl: Inappropriate ioctl
762 for device) when using a non Linux remote box. Thanks to
763 many users and especially to Eros Albertazzi
764 <eros@lamel.bo.cnr.it> for reporting this.
766 7. Found a workaroung for the dump deadlock problem (3 childs
767 stuck in pause(), father in read()). The workaround seems
768 to work for me and several beta-testers. If it doesn't work
769 for you, please report back.
771 8. Updated the RPM spec file (BuildPrereq, URL etc).
773 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
774 ========================================================================
776 1. Fixed the permissions of a newly created QFA file by dump.
778 2. Cleaned up the source of dump (the external variables
779 definition was a complete mess, making possible to have
780 objects overlap).
782 3. Fixed restore to use the full tape volume path when doing
783 a compare (since it changes the working directory to the
784 filesystem being compared in the process).
786 4. Added the -q option to dump which makes dump abort
787 whenever it needs operator attention. It should be
788 easier to use dump in scripts with this option.
790 5. Detect the use of incompatible options to dump and
791 refuse them (like -a and -B options together).
793 6. Added bzip2 compression to dump/restore (use option -j level
794 to select it). Note that you will need a restore version
795 >= 0.4b24 in order to restore a bzip2 compressed dump.
796 The same warning as for the zlib compression applies:
797 the tape format of a bzip2 dump is not compatible with the
798 original BSD tape format.
800 7. Fixed a overflow problem in dump corrupting the dump when
801 very large files were encountered. Thanks to Vladimir Nadvornik
802 for the bug report.
804 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
805 kernel buffer/page cache before starting the dump, helping
806 a bit those who use dump on mounted filesystems. Thanks to
807 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
808 for this suggestion.
810 9. Updated the RPM spec file following the RedHat changes
811 (dynamically linked binaries now in /usr/sbin etc).
813 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
814 which enables restore to recognise multi volume compressed dumps
815 done on CD. Included his scripts for dump (which pipe the dump
816 output directly into cdrecord) and restore. There is now
817 possible to to backups to CD on the fly!
819 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
820 ===================================================================
822 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
823 Marcus Meissner (Caldera International Security Dept.).
825 2. Implement the Sun rmt extended protocol. Patch submitted
826 by Ian Gordon <iangordon@users.sourceforge.net>.
828 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
829 the linux kernel header. This ensures that dump always has the
830 latest version of this file. Patch submitted by
831 Andreas Dilger <adilger@turbolinux.com>.
833 4. Report any filesystem present in either /etc/fstab with a
834 positive passno or /etc/dumpdates in dump -w output.
835 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
837 5. Fixed the looping problem in dump introduced in the
838 previous version.
840 6. Changed the -B option of dump to limit the size of
841 _compressed_ output per volume if compression is on.
842 Patch contributed by Helmut Jarausch
843 <jarausch@igpm.rwth-aachen.de>. Note however that, since
844 it is impossible to predict the size of the compressed
845 data before doing the compression, the -B limit is a bit
846 conservative.
848 7. Fixed a bug in reading the operator typed file/tape path for
849 the next volume in restore.
851 8. Implemented a "-F script" option to restore which permits the
852 user to specify a script which will be launched at the
853 beginning of each tape, useful for automatic programming of
854 tape changers for example. See the restore man page for the
855 script parameters and return codes.
857 9. Small fix for the QFA routines provided by Uwe Gohlke
858 <uwe@ugsoft.de>, and some recommendations for QFA uses in
859 the man pages.
861 10. Fixed the multivolume restoring where making a mistake
862 to the 'Mount next tape' prompt caused several blocks to
863 be lost.
865 11. Enhanced the -e option of dump to take as a parameter a
866 comma separated list of inode numbers.
868 12. Added the -E option to dump which specify a file containing
869 inode numbers to exclude from the dump.
871 13. Fixed the compressed multi-volume dump + restore.
873 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
874 ==================================================================
876 1. Made dump report the number of blocks dumped per volume.
877 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
879 2. Fix a bug in dump where waiting too much at the 'change volumes'
880 question caused the volume to be rewritten. Thanks to
881 Graham Reed <greed@users.sourceforge.net> for reporting the
882 bug and providing a patch.
884 3. Added a compression option to dump, contributed by Jerry
885 Peters <gapeters@worldnet.att.net>.
887 WARNING: the compression works only on files/pipes and
888 drives supporting variable block size.
890 WARNING2: tapes created using the compression feature are
891 incompatible with the standard dump tape format, and a
892 version of restore >= 0.4b22 will be needed for extraction.
894 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
895 architectures. Thanks to Paul Slootman <paul@debian.org> for
896 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
897 it upstream.
899 5. Many cleanups (CPP defines, const char warnings, check of
900 ext2fs COMPAT flags, time_t cleanups, added libext2 version
901 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
903 6. Made --prefix option work in configure. All the install path
904 are now based on the configure parameters.
906 7. Added the Quick File Access mode in dump/restore, contributed
907 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
908 a file tape position for each inode, and this file is used by
909 restore (if called with parameter Q and the filename)
910 to directly position the tape at the file restore is currently
911 working on. This saves hours when restoring single files from
912 large backups, saves the tapes and the drive's head. Use
913 --enable-qfa option of configure to compile in the QFA support.
915 8. Added the possibility to dump several files and directories
916 in a single invocation of dump. Thanks to Uwe Gohlke
917 <uwe@ugsoft.de> for implementing this option.
919 9. Fixed the dumping and restoring of files with holes
920 (files ending with holes didn't get dumped or restored
921 correctly).
923 10. Fixed a socket descriptor leak in dump, which leaved opened
924 3 file descriptors per dump process (and there is one dump
925 process per tape).
927 11. Fixed dump large file system support, by explicit use of
928 open64/lseek64/etc functions (explicit use needed because
929 e2fsprogs libraries don't behave well when compiled with
932 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
933 ======================================================================
935 1. Fixed some bugs in the dump -w|-W logic introduced by
936 the previous version. Thanks to Andreas Dilger
937 <adilger@turbolinux.com> for his help on this one.
939 2. Fixed again a compilation problem when using old e2fs
940 headers (filesystem label related). Thanks to many users
941 who reported this stupid error.
943 3. Fixed a build problem on old lib5 based systems dealing with
944 _PATH_MNTTAB being undefined. Thanks to John Adams
945 <johna@onevista.com> for reporting the bug.
947 4. Improved the error detection in restore when creating
948 the temporary files in TMPDIR. Now restore will corectly
949 report a 'no space left on device' error instead of
950 strange errors which could imply an archive corruption.
951 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
952 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
954 5. Added the throughput information in dump's progression
955 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
956 for the patch.
958 6. Use libext2fs's inode scanning routines, which are particularly
959 robust when dealing with errors (especially when having some
960 bad blocks on the filesystem). Many thanks to Colin
961 <colin@horizon.com> for reporting the bug, investigating
962 the issues, finding a workaround, writing the patch and
963 fully testing it... (of course, if this patch will break
964 anything, he is to blame for :-)).
966 7. Made dump and restore LFS aware. Dump can dump a filesystem
967 containing large files, generate a large file on output and
968 restore can restore them. This feature is not enabled by
969 default, use --enable-largefile option of configure to enable
970 it (you need to have a LFS aware glibc though). Thanks to
971 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
972 and to Theodore T'so <tytso@valinux.com> for his always
973 useful thoughts.
975 8. Made dump ask upon a tape write error if it should rewrite
976 the current volume (assume this is a bad tape) or if it should
977 assume an end-of-tape condition (useful for tape drives which
978 behaves badly at the end of the tape). Thanks to Andreas
979 Dilger <adilger@turbolinux.com> for the suggestion.
981 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
982 =======================================================================
984 1. Fixed a small compilation problem due to a change
985 in the definintion of the struct sigaction in
986 glibc 2.0 / libc5. Thanks to Gunther Schlegel
987 <schlegel@riege.de> for reporting the bug and to
988 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
989 a fix.
991 2. Modified the label and uuid analysis in order to be
992 self-contained (independant of kernel/libc headers). This
993 should fix the compile with older kernel/libc headers and
994 will preserve the functionality. Thanks to Bernhard Erdmann
995 <bernhard.erdmann@gmx.de> for reporting the bug.
997 3. The 'exclude inode' option, if applied to a directory
998 inode, excludes now correctly all the leaves of this
999 directory. Thanks to John R. Dennison
1000 <gerdesas@users.sourceforge.net> for reporting the bug.
1002 4. Fixed the '-e' option to disable the possibility
1003 to exclude the root inode (which causes the dumps to
1004 be unreadable by restore). Prevented array overflow
1005 when multiple -e options are used.
1007 5. Fixed dump to correctly interpret a filesystem argument
1008 which is a mountpoint and it is not an absolute path
1009 (as specified in the fstab). Thanks to Bernhard R. Erdmann
1010 <be@berdmann.de> for reporting the bug.
1012 6. Made dump able to backup files larger than 2 GB. Note that
1013 dump still doesn't cope with files larger than 4 GB.
1015 7. Restore the real uid and gid before invoking an external
1016 RSH program (big hole when dump or restore is suid root!).
1018 8. Get the values for the default tape device and the location
1019 of fstab file from the system headers. Thanks to
1020 Andreas Dilger <adilger@turbolinux.com> for the patch.
1022 9. Made dump -w|-W report all recognized filesystems
1023 present in either /etc/fstab or /etc/dumpdates, and present
1024 the list in the same order as in fstab file. Thanks
1025 to Andreas Dilger <adilger@turbolinux.com> for the patch.
1027 10. Made dump's -a (automatic end of tape detection) the
1028 default. Specifying one of -B, -c, -d or -s options will
1029 override the EOM detection. Thanks to Andreas Dilger
1030 <adilger@turbolinux.com> for the patch.
1032 11. Save the ext2 filesystem label into the dump volume label.
1033 Specifying a volume label on the command line (-L option)
1034 will override this feature. Thanks to Andreas Dilger
1035 <adilger@turbolinux.com> for the patch.
1037 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
1038 =====================================================================
1040 1. Fixed the signal handling in dump (which I broke in 0.4b17)
1041 which was causing several strange dump failures (dump
1042 hanged or segmentation faults etc).
1044 2. Specified the default blocksize in the dump man page.
1046 3. Changed two info messages of restore to be written on stdout
1047 instead of stderr in order to leave stderr only for errors
1048 or warnings. Thanks to Stephen Atwell
1049 <satwell@urbana.css.mot.com> for the suggestion.
1051 4. Corrected an off by one calculation which prevented
1052 dumping a directory having a 1 character name length.
1053 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
1054 for reporting the bug.
1056 5. Reinforce dump in order to allow only level 0 and no
1057 -u option when dumping a subdirectory, and document
1058 this in the man page. Thanks to Bernhard Erdmann
1059 <bernhard.erdmann@gmx.de> for reporting the bug.
1061 6. Fixed a small harmless bug in restore which caused
1062 in some conditions hard links to be restored several
1063 times (therefore generation some warning messages).
1064 Thanks to Kenneth Porter <shiva@well.com> for
1065 reporting the bug.
1067 7. Updated the RPM spec file to the latest RedHat version,
1068 providing FHS packaging and other cosmetic changes.
1069 You will need RPM version >= 3.0.5 in order to build the RPM.
1071 8. Updated the configure script to check for libtermcap
1072 before checking for libreadline, because we need this
1073 library in order to compile the readline support.
1075 9. Made dump understand the LABEL= and UUID= notation
1076 both in /etc/fstab and on the command line. Note that
1077 you will need the /proc filesystem in order to use
1078 these notations. Thanks to Erik Troan <ewt@redhat.com>
1079 for providing the patch.
1081 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
1082 ===================================================================
1084 1. Fixed a potential buffer overflow in restore. Thanks
1085 to Stan Bubrouski <satan@fastdial.net> for reporting
1086 the bug.
1088 2. Fixed a readline-related bug which prevented
1089 'cat DUMPFILE | restore -i -f -' from working. Thanks
1090 to Charles Karney <karney@users.sourceforge.net>
1091 for the bug report.
1093 3. Changed a few "panic" into "exit", causing restore to
1094 be more stable against some attacks (like the last one
1095 reported on Bugtraq, although the last version of restore
1096 was not vulnerable - just dumped core). Thanks to
1097 Andreas Hasenack <andreas@conectiva.com.br> for reporting
1098 the bugs to me.
1100 4. Removed the suid-root bit on dump and restore in the
1101 default build (and generated RPMs). It should be safer
1102 now. Users who need the suid-root capabilities in order
1103 to make network backups should read first the man page
1104 of dump and enable the suid bit manually.
1106 5. Added -ltermcap to the compile parameters for restore
1107 when using readline support, in order to make the compile
1108 process work on some systems (SuSE ?). Thanks to
1109 Patrik Schindler <poc@pocnet.net> for reporting the bug.
1111 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
1112 ====================================================================
1114 1. The -F script is called now *only* at the end of a tape,
1115 not at the end of the dump. Thanks to Christian Haul
1116 <haul@informatik.tu-darmstadt.de> for the bug report.
1118 Normally, the device name and the current volume number
1119 are passed on the command line of the script. However,
1120 if you want the old style script (with no arguments
1121 passed on the command line), you can enable it in
1122 configure with the --enable-oldstylefscript.
1124 2. Use posix signal handling to preserve dump functionality
1125 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
1126 the patch.
1128 3. Made the exit code of restore in case of a 'C'ompare
1129 command reflect the result of the compare. An exit status
1130 of 0 means the dump archive is correct, 1 means tape errors,
1131 2 means that some files were modified. Thanks to Kenneth Porter
1132 <shiva@well.com> for the suggestion.
1134 4. Made (finally) quotes work in interactive restore.
1136 5. Small fixes in order to allow dump to compile with a
1137 really old e2fsprogs version. Thanks to Ian Zimmerman
1138 <itz@speakeasy.org> for the bug report.
1140 6. Add GNU readline capabilities to interactive restore.
1141 Use configure's --enable-readline flag to enable this feature.
1142 Thanks to Patrik Schindler <poc@pocnet.net> for the
1143 suggestion.
1145 7. Do the compare on the fly in restore 'C' mode (this will
1146 allow not to exhaust the available /tmp space when
1147 ccmparing large files). Thanks to Kenneth Porter
1148 <shiva@well.com> for the suggestion.
1150 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
1151 ====================================================================
1153 1. Fixed some several duplicate 'DUMP: DUMP:' in the
1154 output of dump.
1156 2. Corrected the estimation of blocks to dump. Note that
1157 this is still wrong for large directory inodes, due
1158 to the size of a BSD directory entry on the tape
1159 (estimation is based on the size of the ext2 directory,
1160 which is smaller because it doesn't contain the null
1161 character at the end).
1163 3. Made dump report the total number of MB written to
1164 tape at the end of run. Thanks to W. Reilly Cooley
1165 <wcooley@nakedape.cc> for the patch.
1167 4. Added the -X option to restore giving the possibility
1168 to read the names of the files to be extracted/listed
1169 from a text file (in addition of the command line).
1170 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
1171 patch.
1173 5. Added the device name and the current volume number
1174 as arguments to the end of tape script (dump -F option).
1176 6. Made the multi-volume dump work again (after having
1177 broken it in 0.4b15).
1179 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
1180 ===================================================================
1182 1. Added a prompt command in interactive restore mode. Thanks
1183 to Andreas Dilger <adilger@home.com> for the patch.
1185 2. Fixed a buffer overflow problem in dump (caused by
1186 not checking the size of the filesystem parameter).
1187 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
1188 reporting this on Bugtraq (and to several dump users
1189 who forwarded me his mail).
1191 3. Added the '-F script' option to dump in order to
1192 launch a script at the end of each tape (to be used
1193 with a tape changer, or to notify the sysadmin by
1194 pager etc.).
1196 4. Fixed a bug in restore compare code caused by the changes
1197 I made in 0.4b14.
1199 5. Fixed the treatment of options using the old BSD syntax
1200 in both dump and restore.
1202 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
1203 =======================================================================
1205 1. Fixed a bug in dump which may have caused invalid deleted
1206 directories to be dumped out if they were deleted after the
1207 mapping pass. This could occure on active filesystem and lead
1208 to heap corruption (causing dump malfunction in many possible ways).
1209 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
1210 bug and submitting the patch.
1212 2. Fixed the handling of the filesystem parameter in dump. A
1213 '/mnt/disk' parameter caused the disk contents to be dumped,
1214 but a '/mnt/disk/' parameter caused the mountpoint directory
1215 to be dumped (generally an empty directory).
1217 3. Improved the output of dump in order to tell which directory
1218 it is currently dumping (when dumping a subtree).
1220 4. Added the '-e' exclude inode option to dump. Thanks to
1221 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
1223 5. Added a REPORTING-BUGS file in order to provide a guide
1224 on how to correctly report a bug in dump/restore.
1226 6. Fixed a restore problem when restoring a hard link to an inode
1227 having the immutable or append-only attribute set. Thanks to
1228 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
1230 7. Fixed a compatibility problem between dumps made on little
1231 endian machines (the format was correct) and big endian
1232 machines (the format was incorrect). This fix break the
1233 compatibility with the older dumps made on big endian
1234 machines (sparc, m86k, ppc etc). For the first time in
1235 linux dump's history (I believe), the dumps made by *BSD,
1236 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
1237 so interchangeable. Thanks to Rob Cermak
1238 <cermak@ahab.rutgers.edu> for submitting the bug and
1239 helping me test the fix.
1241 8. Fixed the way dump reports the remaining percent/time, if
1242 the number of blocks actually dumped exceeds the estimated
1243 number of blocks. Thanks to Jean-Paul van der Jagt
1244 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
1246 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
1247 ======================================================================
1249 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
1250 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
1252 2. Removed the requirement to build the RPM as root from the
1253 spec file. Thanks to Christian Weisgerber
1254 <naddy@mips.rhein-neckar.de> for submitting this.
1256 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
1257 causing dump to block when dumping really huge filesystems.
1258 Many thanks to Patrik Schindler <poc@pocnet.net> for
1259 helping me find this bug.
1261 4. Fixed the treatment for an interrupt signal when dump access
1262 the remote tape through RSH. Thanks to Christian Weisgerber
1263 <naddy@mips.rhein-neckar.de> for providing the patch.
1265 5. Fixed a bug which was causing dump/restore to display
1266 garbage characters instead of the remote host name.
1268 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
1269 =====================================================================
1271 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
1272 <kukuk@suse.de> for submitting the patch.
1274 2. Fix for the exit code when using the size estimate option of
1275 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
1276 submitting the patch.
1278 3. Handle EINTR in atomical reads/writes in dump, which was causing
1279 dump to fail on some systems. Thanks to Eric Jergensen
1280 <eric@dvns.com> for reporting the bug and submitting the patch.
1282 4. Handle more than 16 characters for the device names in dumpdates.
1283 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
1284 tracking down the problem and proposing the solution.
1286 5. Fixed a bug in dump which prevented the creation of the
1287 dumpdates file when doing a 0-level dump without already
1288 having a dumpdates file. Thanks to Patrik Schindler
1289 <poc@pocnet.net> for reporting the bug.
1291 6. Changed the way dump 'S' flag reports the size estimate
1292 from number of blocks into bytes (making it compatible
1293 with the Solaris version, and simplifying things for
1294 amanda users). Thanks to Jason L Tibbitts III
1295 <tibbs@math.uh.edu> for reporting the bug.
1297 7. Fixed a compatibility problem in linux/alpha dump tape format.
1298 Now the linux/alpha dump are (again) compatible with the
1299 other dump formats. But this breaks compatibility with
1300 older dumps made on alpha. Thanks to Mike Tibor
1301 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
1303 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
1304 ======================================================================
1306 1. Added a '--enable-kerberos' to configure.
1308 2. Added a 'S' option to dump which determines the amount of space
1309 that is needed to perform the dump without actually doing it, similar
1310 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
1311 <cermak@ahab.rutgers.edu>.
1313 3. Added a 'M' multi-volume option to dump and restore which enables
1314 dumping to multiple files (useful when dumping to an ext2
1315 partition to make several dump files in order to bypass the 2GB
1316 file size limitation). The 'f' argument is treated as a prefix and
1317 the output files will be named <prefix>001, <prefix>002 etc. With
1318 the 'M' flag, restore automatically selects the right file without
1319 asking to enter a new tape each time.
1321 4. Fixed a memory leak which was causing dump to grow very big
1322 (270MB when dumping a 10GB filesystem...). Thanks to Jason
1323 Fearon <jasonf@netrider.org.au> for reporting the bug.
1325 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
1326 ======================================================================
1328 1. Make configure test if the system glob routines support
1329 extended syntax (ALTDIRFUNC). If not, use the internal glob
1330 routines instead of system ones. Thanks to Bernhard Sadlowski
1331 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
1332 and helping me resolve this and other minor libc5 compiling
1333 glitches.
1335 2. Fix a problem when dumping a ext2fs with the 'filetype'
1336 feature enabled. Thanks to Patrick J. LoPresti
1337 <patl@cag.lcs.mit.edu> for reporting the bug and to
1338 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
1340 3. Made the nodump flag work on directories. A directory which
1341 has the nodump flag gets never dumped, regardless of its
1342 contents.
1344 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
1345 which allows dump on an active ext3 filesystem. However, this
1346 is a "quick and dirty" patch which enables backup of an ext3
1347 filesystem through the ext2 compatibility (by ignoring the
1348 NEEDS_RECOVERY bit). The journal file is not recognized and
1349 it is dumped (it should not).
1351 5. Test the superblock compatibility flags when dumping, in order
1352 to be sure that we know how to deal with specific features.
1354 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
1355 ====================================================================
1357 1. Use lchown instead of chown, fixing a possible security problem
1358 when restoring symlinks (a malicious user could use this
1359 to deliberately corrupt the ownership of important system files).
1360 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
1361 this and providing the patch.
1363 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
1364 ====================================================================
1366 1. Put dump sources under CVS, added Id tags in all files so
1367 one can use 'ident' on binary files.
1369 2. Added the dump/restore version in the usage text so one can
1370 easily verify the version he is using.
1372 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
1373 a va_start/va_end problem on linux-ppc (always call va_start
1374 va_end in pairs each time we use a vararg function).
1376 4. Added again the DT_* constants because old libc does not
1377 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
1378 for submitting the bug report.
1380 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
1381 this should enable dumping big (huge) filesystems.
1383 6. Added the RSH environment variable in order to be able to
1384 use a rsh replacement like ssh when doing remote backups (and
1385 bypass the security limitations of rcmd). Now you can do remote
1386 backups without being root (or making dump setuid root).
1388 7. Modified again the way dumpdates works. For incremental dumps,
1389 we need to read dumpdates even if we are not using 'u' option.
1390 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
1391 this should work.
1393 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
1394 ===================================================================
1396 1. Removed the 'k' flag from the restore 'about' text if kerberos
1397 was not compiled in.
1399 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
1400 to them (fsetflags takes a char*, setflags an open fd!).
1402 3. (f)setflags is called only if the flags aren't empty. If the
1403 file is a special file, a warning is printed, because changing
1404 flags implies opening the device. Normally, a special file
1405 should not have any flag... (Debian bug #29775, patch provided
1406 by Abhijit Dasgupta <abhijit@ans.net>).
1408 4. Made possible to dump a file system not mentioned in /etc/fstab.
1409 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
1411 5. Changed the default behaviour to not create dumpdates
1412 unless 'u' option is specified. Removed the old "debian-patch"
1413 which provided the same thing. (Debian bug #38136, #33818).
1415 6. Removed all those dump*announce, since they were getting old...
1417 7. Added warning messages if dumpdates does not exist and
1418 when an inferior level dump does not exist (except for a level 0
1419 dump).
1421 8. Debugged the glob calls in interactive mode: restore used a
1422 dirent struct which was different from the /usr/include/dirent.h
1423 one (this used to work, is it a glibc2 change?), so none of the
1424 compat glob (which used /usr/include/dirent.h) or the system glob
1425 worked. Restore use now the system dirent (and the system
1426 DT_* constants), which are compatible with BSD ones.
1428 9. Added a configure flag (--with-dumpdatespath) to specify
1429 the location of dumpdates file. By default, it is
1430 /etc/dumpdates.
1432 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1433 included the current date/version in man pages.
1435 11. Corrected the estimation of remaining time when
1436 the operator doesn't change the tapes quickly enough. This
1437 was an old bug, I thought I corrected it, and discovered
1438 that in fact it was corrected in two different places, so
1439 the results canceled each other...
1441 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1442 ===================================================================
1444 1. Integrated multiple patches from RedHat, Debian and SuSE:
1446 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1447 isn't specified.
1448 - several fixes in the man pages.
1449 - update the default tape device to /dev/st0.
1450 - many updates for Linux Alpha (byte ordering, size_t etc).
1451 - buffer overruns.
1452 - use environment variable for TMPDIR (instead of /tmp).
1453 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1454 - workaround egcs bug (RedHat bugs #4281 and #2989).
1455 - wire $(OPT) throughout Makefile's.
1457 2. Upgrade the dump revision to 1, making possible to dump filesystems
1458 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1460 3. Fix some compile warnings, prototype all functions.
1462 4. Use glibc err/glob instead of internal compatibility
1463 routines (only if available).
1465 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1466 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1468 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1469 ======================================================================
1471 1. Integrated the changes from FreeBSD-3.1-RELEASE
1472 (mostly bug fixes, buffer overruns, dump has now an "automatic
1473 tape length calculation" flag, dump/restore can use kerberos now
1474 (this is NOT tested), use environment variables for TAPE and
1475 RMT etc.).
1477 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1479 3. Corrected a bug in restore when using 'C' option with multi-volumes
1480 tapes (files splited accros two tapes give "size changed" errors
1481 when comparing).
1483 4. Corrected the long standing bug when dumping multiple tapes.
1484 This works for me, needs further testing.
1486 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1487 ====================================================================
1489 1. Dump now runs correctly on kernels 2.1.x
1490 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1492 Changes between versions 0.4b2 and 0.4b3
1493 ========================================
1495 1. Use realpath() if available
1497 2. Report statistics
1499 Changes between versions 0.4b1 and 0.4b2
1500 ========================================
1502 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1503 the patch)
1505 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1507 3. Fixed configure.in to correctly handle --enable-rmt
1509 Changes between versions 0.3 and 0.4b1
1510 ======================================
1512 1. Integrated the changes from 4.4BSD-Lite2
1514 2. Integrated the patches from Debian and Red Hat
1516 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1518 4. Changed dump to use the Ext2fs library to get block addresses. This
1519 should solve the endianness problem on SparcLinux.
1521 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1522 distribution's one) to use autoconf
1524 6. Fixed a few minor bugs
1526 Changes between versions 0.2e and 0.2f
1527 ======================================
1529 1. Added the creation of named pipes (fifos) in restore.
1531 2. Added the -N flag in the restore manual page.
1533 3. Added the file kernel.patch which contains the llseek() optimization
1534 patch for 1.2.x kernels.
1536 4. Fixed a bug in the restoration of symbolic links: owner and group were
1537 not restored.
1539 5. Integrated some changes from FreeBSD 2.2.
1541 6. Added a call to ftruncate() after restoring each file to restore
1542 correctly files ending by a hole.
1544 Changes between versions 0.2d and 0.2e
1545 ======================================
1547 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1548 with owner = root (0) and group = root/wheel/whatever (0).
1550 Changes between versions 0.2c and 0.2d
1551 ======================================
1553 1. Dump is now able to backup 2GB+ filesystems.
1555 2. Dump and restore can now be linked as static binaries.
1557 Changes between versions 0.2b and 0.2c
1558 ======================================
1560 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1562 Changes between versions 0.2a and 0.2b
1563 ======================================
1565 1. Really fixed the bug that I should have corrected in 0.2a.
1567 2. Enabled optimization again.
1569 Changes between versions 0.2 and 0.2a
1570 =====================================
1572 1. Disabled the optimization during compilation.
1574 Changes between versions 0.1 and 0.2
1575 ====================================
1577 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1578 the fs_type field (actually, I modified the file fstab.c to make it
1579 use the mntent functions).
1581 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1582 made by dump should be readable by the BSD restore and backups made
1583 by the BSD dump should be readable by restore. Unfortunately, this
1584 means that the dump format has changed between version 0.1 and version
1585 0.2 :-(
1587 3. Dump is now able to backup a subtree, it is no longer limited to whole
1588 filesystems like the BSD version.
1590 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1591 than 2 GB.
1593 Changes between versions 0.0 and 0.1
1594 ====================================
1596 1. Now create links rdump and rrestore during the `make install' step.
1598 2. Linux port specific bugs added to the manual pages
1600 3. Incorrect estimation of the number of tapes blocks fixed when doing
1601 incremental backups.
1603 4. Better ls-like format in restore in interactive mode.