1 $Id: CHANGES,v 1.327 2011/06/10 12:17:23 stelian Exp $
3 Changes between versions 0.4b43 and 0.4b44 (released June 10, 2011)
4 ===================================================================
6 1. Recognize ext4 partitions in dump -w commands (Sourceforge
7 bug #3125587, RedHat bug #658890). Thanks to Jan Görig
8 <jgorig@users.sourceforge.net> for reporting the bug and
11 2. Fix an issue with multi volume backups, which breaks
12 restoring of files which are splitted on two or more
13 volumes and are starting on 2nd tape or later (RedHat
16 3. Fix a bug in dump when dumping files with holes. This bug
17 was introduced in 0.4b42 while adding ext4 support.
18 (Sourceforge bug #3133762).
20 4. Fix a bug in restore -P where useless index files for compressed
21 dumps were created (Debian bug #583093). Thanks to
22 Mark Wooding <mdw@distorted.org.uk> for the patch.
24 5. Fix SYS_clone invocation on s390 architectures (Sourceforge
25 bug #3303235). Thanks to Adam Tkac <atkac@redhat.com> for
28 6. Fix static builds. Thanks to Peter Volkov <pva@gentoo.org>
31 7. Fix false negatives in configure zlib test. Thanks to
32 Bear Giles <bgiles@coyotesong.com> for the patch.
34 8. Add quick regression script from Bear Giles <bgiles@coyotesong.com>.
36 9. Fix a bug introduced in 0.4b43 causing restore to crash when
37 asking for a new volume when doing multi-volume restores. Thanks
38 to Patrik Schindler <poc@pocnet.net> for reporting the bug.
40 Changes between versions 0.4b42 and 0.4b43 (released June 11, 2010)
41 ===================================================================
43 1. Fix a bug in dump making impossible to handle large toc files
44 (> 2 GB). Thanks to X DUGi <xdugi@users.sourceforge.net> for
45 reporting the bug (Sourceforge bug #2820629)
47 2. Fix 'restore -x' in multi-volume mode, which caused files being
48 spanned on the first and second volumes to be incorrectly extracted.
49 Thanks to Adam Tkac <atkac@redhat.com> for reporting the bug.
50 (there may be several bug reports about this issue, including
51 Sourceforge bug #2117008, RedHat bugs #232415 and #444958).
53 3. Remove -lselinux -lsepol from the standard libraries list,
54 they shouldn't be needed if selinux is not enabled in dump
55 (this also fixes a build failure when selinux libraries are
58 4. Fix restoration of extended attributes of fifos and device nodes.
59 Thanks to Igor Zhbanov <izh1979@gmail.com> for the bug report.
61 5. Don't attempt to set extended attributes on files that were not
62 extracted in 'restore -N' mode. Thanks to Jan Görig
63 <jgorig@users.sourceforge.net> for the patch.
65 6. Make the dump man page explicit about ext4 support.
67 7. Fix a bug in dump which caused EA entries to be unrecognizable by
68 restore in some corner cases. Add a workaround to restore making
69 it able to properly deal with those corrupted EA entries. Thanks
70 to John Austin <jaustin1@users.sourceforge.net> for reporting
73 8. Fix some issues when restoring a dump which was generated using
74 exclusion patterns (either via -e or via the nodump attribute)
75 (see the Debian bug #574667 for details. Thanks to Frédéric
76 Brière <fbriere@fbriere.net> for the bug report and the associated
79 9. Fix progress/estimated blocks display in dump when doing really
80 huge backups. Thanks to Steve Bonds <sbonds@users.sourceforge.net>
81 for the bug report and tests. (Sourceforge bug #2987758)
83 10. Improve level 1 dump speed by rearanging the "mapdirs" code.
84 Thanks to Andreas Kies <andikies@t-online.de> for finding the
85 bottleneck and sending a patch (Sourceforge bug #2998119).
87 11. Use only ctime in order to compare newness instead of both
88 ctime and mtime. This should both speed up dumping time and
89 correct some side-effect bugs (like dumping files with an mtime
90 in the future in every incremental backup). Thanks to
91 Kieran Clancy <codebeard@users.sourceforge.net> for reporting the
92 bug (Sourceforge bug #2999207).
94 12. Extract dumped UNIX sockets instead of ignoring them.
95 (Sourceforge bug #3007216).
97 13. Compiler warning fixes (mainly warn_unused_result ones).
99 Changes between versions 0.4b41 and 0.4b42 (released June 18, 2009)
100 ===================================================================
102 1. Fix printout of the 'Connection to' message in dump/restore.
103 Thanks to Dale Wiles <dwiles@users.sourceforge.net> for the
104 bug report and its fix.
106 2. Fix dump -w/-W output for filesystems having the last backup
107 level equal to 0. Thanks to Pascal Bouchareine
108 <kalou@users.sourceforge.net> for reporting the bug.
110 3. Enable restore to handle restoring onto a different SELinux
111 policy from the dump. Thanks to Tony Nelson
112 <tony_nelson@users.sourceforge.net> for the patch.
113 (see http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=189845
116 4. Fix EA set failures when restoring immutable files. Thanks
117 to Andrew Kroeger for the patch.
119 5. Fix "mode file too large" errors when restoring huge backups.
120 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
123 6. Add the ability to link against libtinfo library if present,
124 instead of libtermcap or libncurses libraries. Thanks to
125 Adam Tkac <atkac@redhat.com> for the patch.
127 7. When comparing a backup, do not consider to be an error the fact
128 that we encounter files created while dump was in progress.
129 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
132 8. Force '-a' to be the default value when running 'restore -C'.
133 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
136 9. Many improvements to the 'cron_dump_to_disk' example.
137 Thanks to Aaron S. Hawley <ashawley@users.sourceforge.net> for
140 10. Fix the QFA generation when extended attributes are backuped
141 along with the inodes. Prior to this fix, the entries for
142 some inodes may get corrupted (duplicated entries or incorrect
143 ones), making the QFA file unusable for repositionning in
144 restore. Many thanks to Kenneth Porter <shiva@sewingwitch.com>
145 for the bug report and the testing of the fix.
147 11. Fix EA restoration in debug mode (Sourceforge bug #1986216).
149 12. Stefan Auracher <star7@onlinehome.de> noticed that the Tower
150 of Hanoi backup strategy documented in the man page wasn't
151 actually based on the Tower of Hanoi algorithm. The man page
152 was updated accordingly.
154 13. Fix missing level information from dump output when using
155 the default level (RedHat bug #493635). Thanks to Adam Tkac
156 <atkac@redhat.com> for the patch.
158 14. Fix the dump man page to reflect the fact that the default
159 dump level is 0 not 9 (RedHat bug #356121). Thanks to Adam Tkac
160 <atkac@redhat.com> for the patch.
162 15. Fixed a few spelling errors in the man pages (RedHat bug #489853).
163 Thanks to Adam Tkac <atkac@redhat.com> for the patch.
165 16. Fixed restore man page and program usage which incorrectly stated
166 that -P and -A were both allowed in the same invocation (RedHat
167 bug #490627). Thanks to Adam Tkac <atkac@redhat.com> for the patch.
169 17. Use sys_clone under Linux to share I/O contexts between dump
170 processes, thus drastically increasing the performance of dump
171 under CFQ (which is the default I/O scheduler used in a number of
172 distributions). Thanks to Jeff Moyer <jmoyer@redhat.com> for the
175 18. Add (preliminary) ext4 support - thanks to libext2fs which does
176 all the job for us. Thanks to Gertjan van Wingerde
177 <gwingerde@gmail.com> for the patch.
179 Changes between versions 0.4b40 and 0.4b41 (released January 2, 2006)
180 =====================================================================
182 1. Fix restore of dumped Access Control Lists. The previous
183 code for EA works fine, but ACLs needed conversion from the
184 ext2/3 disk format to posix_acl format before restoring.
186 2. Fix some issues with restoration of EA on big endian
189 3. Fix restore when the symtab is over 2GB in size.
191 4. Made the directory hash indexing an optional feature,
192 accessible by the '-H' option of restore, and disable it
195 5. Fixed dump to not include extended attributes information
196 in the toc (archive) file which confused restore -t.
198 Changes between versions 0.4b39 and 0.4b40 (released May 2, 2005)
199 =================================================================
201 1. Changed restore to emit warnings (instead of emitting a fatal
202 error) if a file (or a directory) is unavailable for a
203 comparision (if the user doesn't have the necessary permissions
204 to access it for example). Thanks to Kenneth Porter
205 <shiva@sewingwitch.com> for the bug report.
207 2. Re-done the 'do not save directory entries to non-dumped inodes
208 (excluded from dump)' feature. The previous implementation
209 worked well for excluded directories but not for regular files.
210 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
213 3. Fixed a bug in dump where the tape size was miscalculated when
214 the user used -d/-s to specify the tape characteristics. Thanks
215 to Philip Goisman <goisman@physics.arizona.edu> for reporting
218 4. Fixed another bug introduced in restore with the hashtree
219 implementation. This one caused restore to stop saying
220 "removenode: non-empty directory" in some cases.
222 5. Added support for dumping and restoring ext2/3 extended
223 attributes (EA), like the access control lists (ACL) or
224 the security labels used by SELinux.
226 Changes between versions 0.4b38 and 0.4b39 (released January 21, 2005)
227 ======================================================================
229 1. The newly added dump_on_cd_3 example was buggy, replace it
230 with an updated version from Andrew Basterfield
231 <bob@cemetery.homeunix.org>.
233 2. Made restore to chdir() back into the initial directory when
234 dumping core while aborting a comparision operation. The
235 previous behaviour was to write the corefile at the root of
236 the directory being compared, which could very well be
237 read only and preventing the corefile generation. Thanks
238 to Kenneth Porter <shiva@sewingwitch.com> for the bug report.
240 3. Silenced the failure to call fgetflags() when comparing an
241 entry which has no ext2 attributes (as in lsattr()).
243 4. Fix a brown paper bug in restore -C which broke restore and
244 caused modifications on the filesystem being compared
245 (directories containing a file with the same name as the
246 directory get renamed to RSTTMP...). Thanks to Kenneth Porter
247 <shiva@sewingwitch.com> for finding the bug and helping me
250 5. Made restore -C force the -N flag (no writing allowed on
251 the disk) in order to prevent more bugs like the above one.
253 Changes between versions 0.4b37 and 0.4b38 (released January 7, 2005)
254 =====================================================================
256 1. Fix a couple of troff syntax bugs in the man pages.
257 Thanks to Eric Raymond <esr@thyrsus.com> for the patch.
259 2. Made restore use either libncurses or libtermcap, depending
260 on which one is available at configure time.
262 3. Fixed restore negative size display bug when comparing a
263 dump containing files over 2GB. Thanks to Steve Bonds
264 <sbonds@users.sourceforge.net> for the bug report.
266 4. Do not save directory entries to non-dumped inodes
267 (excluded from dump). This will eliminate the 'missing
268 file' warnings when doing 'restore -C'.
270 5. Fix dump crash when backuping a huge (2TB) filesystem,
271 due to a bogus calculation on the inode map size.
272 Thanks to Kevin B. Haines <K.B.Haines@rl.ac.uk> for
273 submitting the bug and testing the fix.
275 6. Fix a problem in restore where the final \0 in the symbolic
276 link names could have been lost, generating corrupt filenames.
277 Thanks to Kyle Wilson <kyle.wilson@amd.com> for reporting the
280 7. Implemented a hash list for the directory names in restore.
281 The linear list used before caused problems in interactive
282 restores when dealing with directories having thousands of
283 entries. Thanks to Brian Ristuccia <bristuccia@starentnetworks.com>
284 for reporting the bug.
286 8. Improved restore -C, this time including the directory
287 attributes into the comparision.
289 9. Made restore understand tapes containing EA/ACLs (which will
290 be dumped by the next version of dump). In this version
291 extended attributes on the tape are ignored, for full EA/ACL
292 support wait for the next version or try the experimental EA
295 Changes between versions 0.4b36 and 0.4b37 (released July 7, 2004)
296 ==================================================================
298 1. Added the --enable-staticz configure option which enables
299 dump and restore to be linked with static versions of
300 libz and libbz2 (and dynamic versions of all other libraries).
301 This will make Debian users happy, because libz and libbz2
302 were the only needed libraries living in /usr, all the
303 others live in /lib. In case of system emergency, it is
304 better not to have to rely on an extra filesystem.
305 Thanks to Bdale Garbee <bdale@gag.com> for the suggestion.
307 2. Fix compilation on (at least the Linux Debian port to) AMD64.
308 (<ext2fs/ext2_types.h> defines some types (__s64 and __u64)
309 that are also defined by <linux/types.h> (<asm/types.h>) and
312 3. Make dump's reading of the dumpdates file a bit more robust,
313 preventing dump from crashing when the dumpdates file has
314 been modified by hand.
316 4. Fixed some offset calculations in dump code which could
317 lead to "bread lseek errors" on large filesystems. Thanks
318 to Bruce Lowekamp <lowekamp@users.sourceforge.net> for
319 reporting this bug and debugging the issue.
321 5. Made dump use the blkid library when searching for devices
322 by label or uuid instead of dump's own routines.
324 6. Corrected a bug in dump where a wrong LABEL=... line in
325 /etc/fstab could prevent dump from dumping unrelated
326 filesystems. Thanks to Bruce Lowekamp
327 <lowekamp@users.sourceforge.net> for reporting the bug.
329 Changes between versions 0.4b35 and 0.4b36 (released April 21, 2004)
330 ====================================================================
332 1. Fixed dump compilation with old gcc versions. Thanks to
333 Mike Castle <dalgoda@users.sourceforge.net> for the patch.
335 2. Fixed some warnings (howmany, roundup, powerof2 redefined)
336 when compiling against a recent glibc version.
338 3. Fixed a bug in restore preventing the read of a dump tape
339 written with Solaris 7 ufsdump. Thanks to Patrick Higgins
340 <phiggins@transzap.com> for reporting the bug and providing
343 4. Changed dump to enable the creation of volumes of different
344 sizes in a single run (make -B accept a list of values).
345 Patch contributed by Florian Zumbiehl <florz@gmx.de>.
347 5. Use the glibc provided minor() and major() macros instead
348 of our own bitmask implementation. This should be safe for
349 when the major/minor namespace will migrate to 32 bits.
350 Thanks to Zhang Jun <zhangjun@nanjing-fnst.com> for reporting
353 6. Made explicit in the dump man page that dump will not create
354 a remote file, it will only write to an already existing one.
356 7. Another try at making size estimates better again.
358 8. Put back the inconditional running of the end-of-tape
359 script which was changed in 0.434 to be run only when -M
360 or multiple -f were NOT specified. Some users rely on this
361 feature even when it is combined with -M/-f.
363 9. Fixed restore when restoring huge backups (where rstdir...
364 temporary files are over 2GB). Thanks to
365 Raphael Jaffey <rjaffey@artic.edu> for reporting this,
366 debugging the issue and making the patch.
368 10. Made restore understand FreeBSD UFS2 tapes. Thanks to
369 David <vrtin@users.sourceforge.net> for submitting the bug
370 and providing a test case.
372 11. Made dump work with an arbitrary integer level (no more 10
373 levels only limitation). Thanks to Matthew
374 <msvincen@midway.uchicago.edu> for the patch.
376 Changes between versions 0.4b34 and 0.4b35 (released December 21, 2003)
377 =======================================================================
379 1. Added a note in the dump man page saying that the default
380 blocksize can be 32 if -d is used with a high density argument.
381 Thanks to Antonios Christofides <A.Christofides@itia.ntua.gr>
384 2. Fixed configure to correctly understand CPPFLAGS, CFLAGS,
385 LDFLAGS environment variables. Thanks to Arcady Genkin
386 <antipode@users.sourceforge.net> for reporting the bug.
388 3. Made -e/-E options of dump accept an unlimited number of inodes
389 to be excluded, rather than a hardcoded maximum. Thanks to
390 Dietrich Rothe <d-rothe@users.sourceforge.net> for the patch.
392 4. Updated the autoconf system to 2.50. Forced the -D_BSD_SOURCE
393 and -D_USE_BSD_SIGNAL defines in configure in order to solve
394 64bit build problems because quad_t is redifined with a
395 different signature. Thanks to Mike Harris <mharris@redhat.com>
396 for reporting this bug.
398 5. Made restore build on Solaris, making possible to
399 restore Linux's "enhanced" tapes. Thanks to Uwe Gohlke
400 <uwe@ugsoft.de> for the patch.
402 6. Made an extension in the dump tape format capable of saving
403 MacOSX specific inode extensions. Uwe Gohlke <uwe@ugsoft.de>
404 wrote the extension and contributed the restore code back
405 into this codebase. The same extension mechanism will be
406 used in the future to save ACLs...
408 7. Made rmt work correctly with regard to QFA and local/remote
409 files and tapes. The remote access will however work only
410 when the dump provided rmt version is used. If you want to
411 use another rmt server, please do not use the QFA feature.
412 Thanks to Uwe Gohlke <uwe@ugsoft.de> for the patch.
414 Changes between versions 0.4b33 and 0.4b34 (released April 18, 2003)
415 ====================================================================
417 1. Fixed the INSTALL file to reflect the actual install paths.
418 Thanks to David Ronis <ronis@ronispc.chem.mcgill.ca> for
421 2. Fixed the configure script to only check for headers presence
422 instead of trying to compile them. This should fix issues
423 with old build environments. Thanks to Kari Lempiainen
424 <kari@funky.pp.fi> for reporting the bug.
426 3. Fixed restore to correctly ignore sockets when comparing
427 a dump (as socket cannot be properly restored anyway). Thanks
428 to Gunther Reiszig <gunther@mit.edu> for reporting the bug.
430 4. Fixed restore to correctly access the archive file (-A argument)
431 even when using a remote tape. Thanks to Fabrice Bellet
432 <fabrice@bellet.info> for reporting the bug.
434 5. Fixed (again) handling of long (largefile) seeks in rmt.
435 Thanks to Fabrice Bellet <fabrice@bellet.info> for reporting
438 6. Fixed restore corner case when dealing with large block sizes
439 dump is able to write now (-b 1024). Thanks to Fabrice Bellet
440 <fabrice@bellet.info> for reporting the bug.
442 7. Fixed a bug preventing dump to access a filesystem having
443 a label exactly 16 bytes in length. Thanks to <minduni@ti-edu.ch>
444 for reporting the bug.
446 8. Made dump store dump dates using explicit timezones, fixing a
447 problem with incremental dumps if the timezone is modified
448 between the dumps. Thanks to Philippe Troin <phil@fifi.org> for
449 the bug report and the patch.
451 9. Fixed a bug encountered when dumping individual files (not full
452 filesystems or directories) and dangling symbolic links happen
453 to be in the list of files. For as far as dump is concerned,
454 dangling symbolic links are allowed, and are dumped as is.
455 Thanks to Jin-su Ahn <jsahn@ee.snu.ac.kr> for reporting the
456 bug and providing the fix.
458 10. Fixed open and creation modes and permissions for QFA and
459 table-of-contents files in dump and restore. Thanks to
460 Philippe Troin <phil@fifi.org> for the patch.
462 11. Fixed the archive file descriptor handling enabling it to be 0.
463 This can happen in some cases when shell redirections are used.
464 Thanks to Philippe Troin <phil@fifi.org> for the patch.
466 12. Delayed the opening of archive file until after suid had been
467 dropped (fixing a possible security issue if dump is suid).
468 Thanks to Philippe Troin <phil@fifi.org> for the patch.
470 13. Fixed the 'S' command handling in the rmt client part.
471 Thanks to Philippe Troin <phil@fifi.org> for the patch.
473 14. Modified the end-of-tape script handling to print out statistics
474 (and stop the timer) before launching the eot script. Also, the eot
475 script does not get run anymore when using -M (which makes sense) or
476 when multiple tapes are listed on the command line
477 (-f tape0,tape1,tapen) (which also makes sense).
478 Thanks to Philippe Troin <phil@fifi.org> for the patch.
480 15. Relicensed dump/restore under the 'revised' BSD license, as per
481 ftp://ftp.cs.berkeley.edu/ucb/4bsd/README.Impt.License.Change.
483 16. Added LZO compression to dump. This new compression method has
484 the advantage of being super fast, thus not killing tape streaming
485 on slow machines. Thanks to Helmut Jarausch
486 <jarausch@igpm.rwth-aachen.de> for the patch and to
487 Markus Oberhumer <markus@oberhumer.com> for giving special permission
488 to include his miniLZO project (GPL licensed) in dump/restore.
490 17. Some small buffer overruns fixes in rmt. Thanks to Antonomasia
491 <ant@notatla.demon.co.uk> for reporting the bugs.
493 18. Added a special rmt version which can do encryption when writing
494 to tape. Read examples/encrypted_rmt/README for details on
495 how to enable and configure it. Thanks to Ken Lalonde
496 <ken@globalremit.com> for the patch.
498 19. Made dump work with 2.5 kernel end of tape early warning semantics.
499 Thanks to Kai Makisara <Kai.Makisara@kolumbus.fi> for the patch.
501 20. Fixed a bug which caused dump -w|-W not to work anymore, because
502 the fs_freq and fs_passno fields in /etc/mtab are always set
503 to 0 0. Thanks to Trent Piepho <xyzzy@speakeasy.org> for
506 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
507 =======================================================================
509 1. Added a note in the restore man page clarifying the question
510 'set the permissions on the current directory ?' asked by
511 restore at the end of treatment in -i and -x modes.
513 2. Fixed the endianess issues when reading compressed tapes.
514 Thanks to Dark Force <daq4th@users.sourceforge.net> for
515 reporting this bug and providing test cases.
517 3. Fixed the "ACL won't be dumped" warning message (which showed
518 an extra, unrelated error message). Thanks to Dragan Krnic
519 <dkrnic@lycos.com> for reporting this bug.
521 4. Made dump look first into /etc/mtab, then into /etc/fstab
522 when searching for filesystem paths. Also fixed some problems
523 caused by binding mounts or multiple block device mounts.
524 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
525 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
526 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
527 and #160305, Sourceforge bugs #618699 and #679832).
529 5. Made dump's -I option accept the value '0' meaning all the
530 read errors will be ignored. This can be useful when running
531 dump from unattended sessions (like cron jobs). Thanks to
532 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
534 6. Fixed the output of dump to indicate 'blocks' instead of
535 'tape blocks' in the various messages (blocks are always
536 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
537 made it clearly print the current blocksize at the start of
538 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
541 7. Made rmt understand long (largefiles) seeks.
543 8. Fixed build with very old versions of libext2fs, where
544 EXT2_FT_* constants were undefined.
546 9. Made dump accept the dumpdates path on the command line
547 (-D file option) instead of using only the hardcoded one.
548 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
551 10. Enabled rmt, LFS, readline, QFA options by default in
552 ./configure. Updated the configure process (new versions
553 of config.guess, config.sub etc).
555 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
556 =======================================================================
558 1. Changed dump to use fcntl(F_SETLK) style locking instead
559 of flock() when locking the dumpdates file. With the old
560 locking scheme, a local user having read rights on the
561 dumpdates file could be able to do a Denial of Service attack
562 on dump. In order to lock the dumpdates file with the new
563 scheme, the user would need to have write access on the file.
564 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
565 reporting the bug (originally a bugtraq post).
567 2. Fixed interactive 'ls' which caused spurious errors warnings
568 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
569 <jorgen_ostling@users.sourceforge.net> for reporting this
572 3. Fixed dump's estimate when dealing with sparse inodes.
574 4. Modified dump to allow setting a blocksize bigger than 32kB
575 (raised the limit to 1024kB), because newer hardware needs
576 this for proper operation. Thanks to Dirk Traenapp
577 <dtraenapp@users.sourceforge.net> for reporting this.
579 5. Fixed a bug causing Dump to stop and report an error if an
580 inode number in the exclude file was followed by some amount
581 of whitespace. Thanks to Jeffrey Sofferin
582 <sofferin@users.sourceforge.net> for reporting this bug.
584 6. Fixed a bug which caused restore, in some particular cases,
585 to ask some 'scary' questions and leave a bunch of RSTTMP
586 directories behind when restoring incremental tapes. Thanks
587 to Philippe Troin <phil@fifi.org> for reporting this bug and
588 providing the test cases.
590 7. Changed the wording when inodes are excluded from dump:
591 replaced 'Added inode 7 to exclude list' with
592 'Excluding inode 7 (resize inode) from dump', as suggested
593 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
595 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
596 ===================================================================
598 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
599 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
600 <eros@lamel.bo.cnr.it> for reporting the bug.
602 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
603 ===================================================================
605 1. Made dump print out the ext2 logical block number in case of a read
606 error, which can be given as an argument to debugfs' ncheck command,
607 in order to find out the affected inode. Added note about this
608 usage in dump's man page.
610 2. Fixed a problem in restore when reading tapes written on big
611 endian machines with very old versions of dump. The patch was
612 contributed by George Helffrich <george@geology.bristol.ac.uk>.
614 3. Fixed the tape length calculation when using large tapes
615 and compression. Thanks to Georg Lippold
616 <g_lippold@sourceforge.net> for reporting the bug.
618 4. Added a new set of examples in dump_on_cd_2 directory, based
619 on dump_on_cd examples but somewhat enhanced, supporting DVD
620 media, and localized in english and german. Thanks to
621 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
623 5. Made dump save 32 bit UID/GID. Internally, this was achieved
624 by switching from the old BSD inode format to the new BSD
625 inode format, which means that the tape format was changed.
626 However, since all restore versions out there should
627 transparently support both inode formats, the change should
628 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
631 6. Fixed a lot of warnings in the code shown when compiling
632 with 'gcc -W'. Thanks to Matthias Andree
633 <matthias.andree@stud.uni-dortmund.de> for reporting this.
635 7. Fixed a small markup bug in the dump man page. Thanks to
636 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
639 8. Rewrote entirely the man pages using the tmac.an macro
640 package (Linux man page format) instead of the original BSD
641 format. They should be now cleaner and easier to modify.
643 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
644 ==================================================================
646 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
647 non Linux clients were misinterpreted. The description of the problem
648 (incompatible numbering in Linux mtio opcodes) is documented at
649 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
650 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
651 providing an excellent, cross-platform replacement for rmt in his
654 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
655 which caused restore to fail when the inode maps were not entirely
656 contained on the first volume (when using really small volumes or
657 when dumping a huge number of inodes).
659 3. Fixed a problem in dump, where files in subdirectories of directories
660 that have the nodump flag set (or being excluded with -e/-E) were
661 sometimes incorrectly dumped (instead of being excluded). The original
662 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
663 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
664 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
665 help testing the patch.
667 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
668 which makes one able to pipe the output of dump, by the net, to
669 a remote CD-burner server.
671 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
672 and use GNU's symbolic syntax over rmt instead of numerical values
673 to assure multiple platform compatibility.
675 6. Documented the -d option in restore.
677 7. Added a -v (verbose) mode to dump. For now it just prints the number
678 of the inode being dumped, but this could evolve in future versions
679 to include interesting debugging output.
681 8. Added a -o flag to restore, which automatically restores the current
682 directory permissions in -i or -x mode, without asking the operator
683 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
684 and Peter Samuel <peters@e-smith.com>.
686 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
687 ====================================================================
689 1. Fixed a bug in the QFA feature which could have caused
690 a bad tape positionning on restore (causing extra delays in
691 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
693 2. Added a small note in the dump man page specifying that
694 there should be no white space between the option letter and
695 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
697 3. Made restore work with older versions of the readline library,
698 by checking in configure for several extended symbols. Restore
699 can now be compiled with a readline as old as the 2.0 release
700 (though it may be a good idea to upgrade it to a more recent
701 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
702 reporting the build failures.
704 4. Fixed a performance problem with the QFA file creation in
705 dump, which made unnecessary seeks on the tape slowing down
706 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
707 reporting this issue.
709 5. Removed the inclusion of some kernel headers in the dump
710 source, which prevented the compile in some kernel/glibc
711 headers/architecture combination. Thanks to Bdale Garbee
712 <bdale@gag.com> for reporting the bug.
714 6. Added the appropriate error message when dump fails to
715 open the output file for writing. Thanks to Amith Varghese
716 <amithv@yahoo.com> for reporting this bug.
718 7. Made restore able to understand large Solaris ufsdump tapes
719 (containing inodes bigger than 4194304). Sun have introduced
720 an "extension" to the dump tape format when dealing with
721 those inodes, which was uncorrectly handled by Linux restore.
722 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
723 providing a test case.
725 8. Added the -m parameter to dump which optimises the output for
726 inodes having been changed but not modified since the last dump
727 ('changed' and 'modified' have the meaning defined in stat(2)).
728 For those inodes, dump will save only the metadata, instead of
729 saving the entire inode contents. Inodes which are either
730 directories or have been modified since the last dump are saved
731 in a regular way. Uses of this flag must be consistent, meaning
732 that either every dump in an incremental dump set have the flag,
735 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
736 =======================================================================
738 1. Fixed behaviour of dump when exceeding resource limits
741 2. Added the -L flag to restore to allow the user to specify a
742 maximal allowed number of miscompares when using restore
743 with the -C option to check the backup.
745 3. Detailed the manual entry for the -N option of restore.
747 4. Added the -a flag to restore to make able doing unattended
748 restores in -i or -x mode (automatically walks through the
749 multiple dump volumes).
751 5. Extended the QFA mode to work with local files and/or
752 remote tapes and files. This way, restore can know in advance
753 the tape number and the offset for the inodes to extract and
754 can minimize the extraction time by seeking directly to the
757 6. Added the -A <archive> option to both dump and restore,
758 which makes dump to archive a dump table-of-contents in
759 the specified file to be used by restore to determine
760 whether a file is in the dump file that is being restored.
761 (the archive file syntax is also compatible with the
762 Solaris ufsdump generated one).
764 7. Small fix in restore making it able to read some (broken ?)
765 Solaris ufsdump tapes.
767 8. Fixed dump to correctly recognise the root filesystem when using
768 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
769 for reporting this bug.
771 9. Added the -P <file> option to restore to create a
772 Quick File Access file from an already made dump. Patch
773 contributed by Uwe Gohlke <uwe@ugsoft.de>.
775 10. Made restore compile and run on Solaris, making it a
776 possible replacement for the standard ufsrestore. Port was
777 contributed by Uwe Gohlke <uwe@ugsoft.de>.
779 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
780 =====================================================================
782 1. Added a set of backup scripts from Eugenio Diaz
783 <getnito@yahoo.com> in the examples section. It features
784 automatic (cron based) full and incremental dumping of
785 several filesystems on a separate filesystem.
787 2. Fixed a off-by-one miscalculation which disabled dumping
788 a one letter subdirectory. Thanks to Chris Siebenmann
789 <cks@utcc.utoronto.ca> for reporting the bug.
791 3. Fixed several restore bugs occuring when trying to
792 restore some missing files on the tape. Thanks to Chris
793 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
795 4. Fixed --with-ldopts configure argument passing, installing from
796 a separate object directory, makefile cleanups contributed
797 by <splite@purdue.edu>.
799 5. Fix a bug which could caused, in some conditions, the highest
800 number inode of a filesystem, to not be dumped. Many thanks
801 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
804 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
805 =======================================================================
807 1. Added a mini howto from Patrick Walsh in the examples
810 2. Minor man pages syntax corrections. Thanks to
811 Chris Danis <screechco@home.com> for reporting the bugs.
813 3. Added a script from David B. Peterson <dave@toppledwagon.com>
814 to the examples section. It features dumping several
815 filesystems to a remote tape drive upon ssh.
817 4. Added a patch provided by Richard Jones <rich@annexia.org>
818 which allows BRADEMAX (number of read errors tolerated by
819 dump) to be adjusted using the -I option.
821 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
822 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
825 6. Add the -l option to restore to specify if, when doing a
826 remote restore, the file used to access the backup is a
827 regular file (the defaults being a tape drive). Restore needs
828 to know this information when reading a remote compressed
829 dump. Previously, this information was autodetected, but
830 the autodetection code fails (with ioctl: Inappropriate ioctl
831 for device) when using a non Linux remote box. Thanks to
832 many users and especially to Eros Albertazzi
833 <eros@lamel.bo.cnr.it> for reporting this.
835 7. Found a workaroung for the dump deadlock problem (3 childs
836 stuck in pause(), father in read()). The workaround seems
837 to work for me and several beta-testers. If it doesn't work
838 for you, please report back.
840 8. Updated the RPM spec file (BuildPrereq, URL etc).
842 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
843 ========================================================================
845 1. Fixed the permissions of a newly created QFA file by dump.
847 2. Cleaned up the source of dump (the external variables
848 definition was a complete mess, making possible to have
851 3. Fixed restore to use the full tape volume path when doing
852 a compare (since it changes the working directory to the
853 filesystem being compared in the process).
855 4. Added the -q option to dump which makes dump abort
856 whenever it needs operator attention. It should be
857 easier to use dump in scripts with this option.
859 5. Detect the use of incompatible options to dump and
860 refuse them (like -a and -B options together).
862 6. Added bzip2 compression to dump/restore (use option -j level
863 to select it). Note that you will need a restore version
864 >= 0.4b24 in order to restore a bzip2 compressed dump.
865 The same warning as for the zlib compression applies:
866 the tape format of a bzip2 dump is not compatible with the
867 original BSD tape format.
869 7. Fixed a overflow problem in dump corrupting the dump when
870 very large files were encountered. Thanks to Vladimir Nadvornik
873 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
874 kernel buffer/page cache before starting the dump, helping
875 a bit those who use dump on mounted filesystems. Thanks to
876 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
879 9. Updated the RPM spec file following the RedHat changes
880 (dynamically linked binaries now in /usr/sbin etc).
882 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
883 which enables restore to recognise multi volume compressed dumps
884 done on CD. Included his scripts for dump (which pipe the dump
885 output directly into cdrecord) and restore. There is now
886 possible to to backups to CD on the fly!
888 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
889 ===================================================================
891 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
892 Marcus Meissner (Caldera International Security Dept.).
894 2. Implement the Sun rmt extended protocol. Patch submitted
895 by Ian Gordon <iangordon@users.sourceforge.net>.
897 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
898 the linux kernel header. This ensures that dump always has the
899 latest version of this file. Patch submitted by
900 Andreas Dilger <adilger@turbolinux.com>.
902 4. Report any filesystem present in either /etc/fstab with a
903 positive passno or /etc/dumpdates in dump -w output.
904 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
906 5. Fixed the looping problem in dump introduced in the
909 6. Changed the -B option of dump to limit the size of
910 _compressed_ output per volume if compression is on.
911 Patch contributed by Helmut Jarausch
912 <jarausch@igpm.rwth-aachen.de>. Note however that, since
913 it is impossible to predict the size of the compressed
914 data before doing the compression, the -B limit is a bit
917 7. Fixed a bug in reading the operator typed file/tape path for
918 the next volume in restore.
920 8. Implemented a "-F script" option to restore which permits the
921 user to specify a script which will be launched at the
922 beginning of each tape, useful for automatic programming of
923 tape changers for example. See the restore man page for the
924 script parameters and return codes.
926 9. Small fix for the QFA routines provided by Uwe Gohlke
927 <uwe@ugsoft.de>, and some recommendations for QFA uses in
930 10. Fixed the multivolume restoring where making a mistake
931 to the 'Mount next tape' prompt caused several blocks to
934 11. Enhanced the -e option of dump to take as a parameter a
935 comma separated list of inode numbers.
937 12. Added the -E option to dump which specify a file containing
938 inode numbers to exclude from the dump.
940 13. Fixed the compressed multi-volume dump + restore.
942 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
943 ==================================================================
945 1. Made dump report the number of blocks dumped per volume.
946 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
948 2. Fix a bug in dump where waiting too much at the 'change volumes'
949 question caused the volume to be rewritten. Thanks to
950 Graham Reed <greed@users.sourceforge.net> for reporting the
951 bug and providing a patch.
953 3. Added a compression option to dump, contributed by Jerry
954 Peters <gapeters@worldnet.att.net>.
956 WARNING: the compression works only on files/pipes and
957 drives supporting variable block size.
959 WARNING2: tapes created using the compression feature are
960 incompatible with the standard dump tape format, and a
961 version of restore >= 0.4b22 will be needed for extraction.
963 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
964 architectures. Thanks to Paul Slootman <paul@debian.org> for
965 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
968 5. Many cleanups (CPP defines, const char warnings, check of
969 ext2fs COMPAT flags, time_t cleanups, added libext2 version
970 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
972 6. Made --prefix option work in configure. All the install path
973 are now based on the configure parameters.
975 7. Added the Quick File Access mode in dump/restore, contributed
976 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
977 a file tape position for each inode, and this file is used by
978 restore (if called with parameter Q and the filename)
979 to directly position the tape at the file restore is currently
980 working on. This saves hours when restoring single files from
981 large backups, saves the tapes and the drive's head. Use
982 --enable-qfa option of configure to compile in the QFA support.
984 8. Added the possibility to dump several files and directories
985 in a single invocation of dump. Thanks to Uwe Gohlke
986 <uwe@ugsoft.de> for implementing this option.
988 9. Fixed the dumping and restoring of files with holes
989 (files ending with holes didn't get dumped or restored
992 10. Fixed a socket descriptor leak in dump, which leaved opened
993 3 file descriptors per dump process (and there is one dump
996 11. Fixed dump large file system support, by explicit use of
997 open64/lseek64/etc functions (explicit use needed because
998 e2fsprogs libraries don't behave well when compiled with
999 FILE_OFFSET_BITS=64).
1001 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
1002 ======================================================================
1004 1. Fixed some bugs in the dump -w|-W logic introduced by
1005 the previous version. Thanks to Andreas Dilger
1006 <adilger@turbolinux.com> for his help on this one.
1008 2. Fixed again a compilation problem when using old e2fs
1009 headers (filesystem label related). Thanks to many users
1010 who reported this stupid error.
1012 3. Fixed a build problem on old lib5 based systems dealing with
1013 _PATH_MNTTAB being undefined. Thanks to John Adams
1014 <johna@onevista.com> for reporting the bug.
1016 4. Improved the error detection in restore when creating
1017 the temporary files in TMPDIR. Now restore will corectly
1018 report a 'no space left on device' error instead of
1019 strange errors which could imply an archive corruption.
1020 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
1021 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
1023 5. Added the throughput information in dump's progression
1024 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
1027 6. Use libext2fs's inode scanning routines, which are particularly
1028 robust when dealing with errors (especially when having some
1029 bad blocks on the filesystem). Many thanks to Colin
1030 <colin@horizon.com> for reporting the bug, investigating
1031 the issues, finding a workaround, writing the patch and
1032 fully testing it... (of course, if this patch will break
1033 anything, he is to blame for :-)).
1035 7. Made dump and restore LFS aware. Dump can dump a filesystem
1036 containing large files, generate a large file on output and
1037 restore can restore them. This feature is not enabled by
1038 default, use --enable-largefile option of configure to enable
1039 it (you need to have a LFS aware glibc though). Thanks to
1040 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
1041 and to Theodore T'so <tytso@valinux.com> for his always
1044 8. Made dump ask upon a tape write error if it should rewrite
1045 the current volume (assume this is a bad tape) or if it should
1046 assume an end-of-tape condition (useful for tape drives which
1047 behaves badly at the end of the tape). Thanks to Andreas
1048 Dilger <adilger@turbolinux.com> for the suggestion.
1050 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
1051 =======================================================================
1053 1. Fixed a small compilation problem due to a change
1054 in the definintion of the struct sigaction in
1055 glibc 2.0 / libc5. Thanks to Gunther Schlegel
1056 <schlegel@riege.de> for reporting the bug and to
1057 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
1060 2. Modified the label and uuid analysis in order to be
1061 self-contained (independant of kernel/libc headers). This
1062 should fix the compile with older kernel/libc headers and
1063 will preserve the functionality. Thanks to Bernhard Erdmann
1064 <bernhard.erdmann@gmx.de> for reporting the bug.
1066 3. The 'exclude inode' option, if applied to a directory
1067 inode, excludes now correctly all the leaves of this
1068 directory. Thanks to John R. Dennison
1069 <gerdesas@users.sourceforge.net> for reporting the bug.
1071 4. Fixed the '-e' option to disable the possibility
1072 to exclude the root inode (which causes the dumps to
1073 be unreadable by restore). Prevented array overflow
1074 when multiple -e options are used.
1076 5. Fixed dump to correctly interpret a filesystem argument
1077 which is a mountpoint and it is not an absolute path
1078 (as specified in the fstab). Thanks to Bernhard R. Erdmann
1079 <be@berdmann.de> for reporting the bug.
1081 6. Made dump able to backup files larger than 2 GB. Note that
1082 dump still doesn't cope with files larger than 4 GB.
1084 7. Restore the real uid and gid before invoking an external
1085 RSH program (big hole when dump or restore is suid root!).
1087 8. Get the values for the default tape device and the location
1088 of fstab file from the system headers. Thanks to
1089 Andreas Dilger <adilger@turbolinux.com> for the patch.
1091 9. Made dump -w|-W report all recognized filesystems
1092 present in either /etc/fstab or /etc/dumpdates, and present
1093 the list in the same order as in fstab file. Thanks
1094 to Andreas Dilger <adilger@turbolinux.com> for the patch.
1096 10. Made dump's -a (automatic end of tape detection) the
1097 default. Specifying one of -B, -c, -d or -s options will
1098 override the EOM detection. Thanks to Andreas Dilger
1099 <adilger@turbolinux.com> for the patch.
1101 11. Save the ext2 filesystem label into the dump volume label.
1102 Specifying a volume label on the command line (-L option)
1103 will override this feature. Thanks to Andreas Dilger
1104 <adilger@turbolinux.com> for the patch.
1106 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
1107 =====================================================================
1109 1. Fixed the signal handling in dump (which I broke in 0.4b17)
1110 which was causing several strange dump failures (dump
1111 hanged or segmentation faults etc).
1113 2. Specified the default blocksize in the dump man page.
1115 3. Changed two info messages of restore to be written on stdout
1116 instead of stderr in order to leave stderr only for errors
1117 or warnings. Thanks to Stephen Atwell
1118 <satwell@urbana.css.mot.com> for the suggestion.
1120 4. Corrected an off by one calculation which prevented
1121 dumping a directory having a 1 character name length.
1122 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
1123 for reporting the bug.
1125 5. Reinforce dump in order to allow only level 0 and no
1126 -u option when dumping a subdirectory, and document
1127 this in the man page. Thanks to Bernhard Erdmann
1128 <bernhard.erdmann@gmx.de> for reporting the bug.
1130 6. Fixed a small harmless bug in restore which caused
1131 in some conditions hard links to be restored several
1132 times (therefore generation some warning messages).
1133 Thanks to Kenneth Porter <shiva@well.com> for
1136 7. Updated the RPM spec file to the latest RedHat version,
1137 providing FHS packaging and other cosmetic changes.
1138 You will need RPM version >= 3.0.5 in order to build the RPM.
1140 8. Updated the configure script to check for libtermcap
1141 before checking for libreadline, because we need this
1142 library in order to compile the readline support.
1144 9. Made dump understand the LABEL= and UUID= notation
1145 both in /etc/fstab and on the command line. Note that
1146 you will need the /proc filesystem in order to use
1147 these notations. Thanks to Erik Troan <ewt@redhat.com>
1148 for providing the patch.
1150 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
1151 ===================================================================
1153 1. Fixed a potential buffer overflow in restore. Thanks
1154 to Stan Bubrouski <satan@fastdial.net> for reporting
1157 2. Fixed a readline-related bug which prevented
1158 'cat DUMPFILE | restore -i -f -' from working. Thanks
1159 to Charles Karney <karney@users.sourceforge.net>
1162 3. Changed a few "panic" into "exit", causing restore to
1163 be more stable against some attacks (like the last one
1164 reported on Bugtraq, although the last version of restore
1165 was not vulnerable - just dumped core). Thanks to
1166 Andreas Hasenack <andreas@conectiva.com.br> for reporting
1169 4. Removed the suid-root bit on dump and restore in the
1170 default build (and generated RPMs). It should be safer
1171 now. Users who need the suid-root capabilities in order
1172 to make network backups should read first the man page
1173 of dump and enable the suid bit manually.
1175 5. Added -ltermcap to the compile parameters for restore
1176 when using readline support, in order to make the compile
1177 process work on some systems (SuSE ?). Thanks to
1178 Patrik Schindler <poc@pocnet.net> for reporting the bug.
1180 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
1181 ====================================================================
1183 1. The -F script is called now *only* at the end of a tape,
1184 not at the end of the dump. Thanks to Christian Haul
1185 <haul@informatik.tu-darmstadt.de> for the bug report.
1187 Normally, the device name and the current volume number
1188 are passed on the command line of the script. However,
1189 if you want the old style script (with no arguments
1190 passed on the command line), you can enable it in
1191 configure with the --enable-oldstylefscript.
1193 2. Use posix signal handling to preserve dump functionality
1194 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
1197 3. Made the exit code of restore in case of a 'C'ompare
1198 command reflect the result of the compare. An exit status
1199 of 0 means the dump archive is correct, 1 means tape errors,
1200 2 means that some files were modified. Thanks to Kenneth Porter
1201 <shiva@well.com> for the suggestion.
1203 4. Made (finally) quotes work in interactive restore.
1205 5. Small fixes in order to allow dump to compile with a
1206 really old e2fsprogs version. Thanks to Ian Zimmerman
1207 <itz@speakeasy.org> for the bug report.
1209 6. Add GNU readline capabilities to interactive restore.
1210 Use configure's --enable-readline flag to enable this feature.
1211 Thanks to Patrik Schindler <poc@pocnet.net> for the
1214 7. Do the compare on the fly in restore 'C' mode (this will
1215 allow not to exhaust the available /tmp space when
1216 ccmparing large files). Thanks to Kenneth Porter
1217 <shiva@well.com> for the suggestion.
1219 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
1220 ====================================================================
1222 1. Fixed some several duplicate 'DUMP: DUMP:' in the
1225 2. Corrected the estimation of blocks to dump. Note that
1226 this is still wrong for large directory inodes, due
1227 to the size of a BSD directory entry on the tape
1228 (estimation is based on the size of the ext2 directory,
1229 which is smaller because it doesn't contain the null
1230 character at the end).
1232 3. Made dump report the total number of MB written to
1233 tape at the end of run. Thanks to W. Reilly Cooley
1234 <wcooley@nakedape.cc> for the patch.
1236 4. Added the -X option to restore giving the possibility
1237 to read the names of the files to be extracted/listed
1238 from a text file (in addition of the command line).
1239 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
1242 5. Added the device name and the current volume number
1243 as arguments to the end of tape script (dump -F option).
1245 6. Made the multi-volume dump work again (after having
1246 broken it in 0.4b15).
1248 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
1249 ===================================================================
1251 1. Added a prompt command in interactive restore mode. Thanks
1252 to Andreas Dilger <adilger@home.com> for the patch.
1254 2. Fixed a buffer overflow problem in dump (caused by
1255 not checking the size of the filesystem parameter).
1256 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
1257 reporting this on Bugtraq (and to several dump users
1258 who forwarded me his mail).
1260 3. Added the '-F script' option to dump in order to
1261 launch a script at the end of each tape (to be used
1262 with a tape changer, or to notify the sysadmin by
1265 4. Fixed a bug in restore compare code caused by the changes
1268 5. Fixed the treatment of options using the old BSD syntax
1269 in both dump and restore.
1271 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
1272 =======================================================================
1274 1. Fixed a bug in dump which may have caused invalid deleted
1275 directories to be dumped out if they were deleted after the
1276 mapping pass. This could occure on active filesystem and lead
1277 to heap corruption (causing dump malfunction in many possible ways).
1278 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
1279 bug and submitting the patch.
1281 2. Fixed the handling of the filesystem parameter in dump. A
1282 '/mnt/disk' parameter caused the disk contents to be dumped,
1283 but a '/mnt/disk/' parameter caused the mountpoint directory
1284 to be dumped (generally an empty directory).
1286 3. Improved the output of dump in order to tell which directory
1287 it is currently dumping (when dumping a subtree).
1289 4. Added the '-e' exclude inode option to dump. Thanks to
1290 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
1292 5. Added a REPORTING-BUGS file in order to provide a guide
1293 on how to correctly report a bug in dump/restore.
1295 6. Fixed a restore problem when restoring a hard link to an inode
1296 having the immutable or append-only attribute set. Thanks to
1297 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
1299 7. Fixed a compatibility problem between dumps made on little
1300 endian machines (the format was correct) and big endian
1301 machines (the format was incorrect). This fix break the
1302 compatibility with the older dumps made on big endian
1303 machines (sparc, m86k, ppc etc). For the first time in
1304 linux dump's history (I believe), the dumps made by *BSD,
1305 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
1306 so interchangeable. Thanks to Rob Cermak
1307 <cermak@ahab.rutgers.edu> for submitting the bug and
1308 helping me test the fix.
1310 8. Fixed the way dump reports the remaining percent/time, if
1311 the number of blocks actually dumped exceeds the estimated
1312 number of blocks. Thanks to Jean-Paul van der Jagt
1313 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
1315 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
1316 ======================================================================
1318 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
1319 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
1321 2. Removed the requirement to build the RPM as root from the
1322 spec file. Thanks to Christian Weisgerber
1323 <naddy@mips.rhein-neckar.de> for submitting this.
1325 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
1326 causing dump to block when dumping really huge filesystems.
1327 Many thanks to Patrik Schindler <poc@pocnet.net> for
1328 helping me find this bug.
1330 4. Fixed the treatment for an interrupt signal when dump access
1331 the remote tape through RSH. Thanks to Christian Weisgerber
1332 <naddy@mips.rhein-neckar.de> for providing the patch.
1334 5. Fixed a bug which was causing dump/restore to display
1335 garbage characters instead of the remote host name.
1337 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
1338 =====================================================================
1340 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
1341 <kukuk@suse.de> for submitting the patch.
1343 2. Fix for the exit code when using the size estimate option of
1344 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
1345 submitting the patch.
1347 3. Handle EINTR in atomical reads/writes in dump, which was causing
1348 dump to fail on some systems. Thanks to Eric Jergensen
1349 <eric@dvns.com> for reporting the bug and submitting the patch.
1351 4. Handle more than 16 characters for the device names in dumpdates.
1352 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
1353 tracking down the problem and proposing the solution.
1355 5. Fixed a bug in dump which prevented the creation of the
1356 dumpdates file when doing a 0-level dump without already
1357 having a dumpdates file. Thanks to Patrik Schindler
1358 <poc@pocnet.net> for reporting the bug.
1360 6. Changed the way dump 'S' flag reports the size estimate
1361 from number of blocks into bytes (making it compatible
1362 with the Solaris version, and simplifying things for
1363 amanda users). Thanks to Jason L Tibbitts III
1364 <tibbs@math.uh.edu> for reporting the bug.
1366 7. Fixed a compatibility problem in linux/alpha dump tape format.
1367 Now the linux/alpha dump are (again) compatible with the
1368 other dump formats. But this breaks compatibility with
1369 older dumps made on alpha. Thanks to Mike Tibor
1370 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
1372 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
1373 ======================================================================
1375 1. Added a '--enable-kerberos' to configure.
1377 2. Added a 'S' option to dump which determines the amount of space
1378 that is needed to perform the dump without actually doing it, similar
1379 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
1380 <cermak@ahab.rutgers.edu>.
1382 3. Added a 'M' multi-volume option to dump and restore which enables
1383 dumping to multiple files (useful when dumping to an ext2
1384 partition to make several dump files in order to bypass the 2GB
1385 file size limitation). The 'f' argument is treated as a prefix and
1386 the output files will be named <prefix>001, <prefix>002 etc. With
1387 the 'M' flag, restore automatically selects the right file without
1388 asking to enter a new tape each time.
1390 4. Fixed a memory leak which was causing dump to grow very big
1391 (270MB when dumping a 10GB filesystem...). Thanks to Jason
1392 Fearon <jasonf@netrider.org.au> for reporting the bug.
1394 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
1395 ======================================================================
1397 1. Make configure test if the system glob routines support
1398 extended syntax (ALTDIRFUNC). If not, use the internal glob
1399 routines instead of system ones. Thanks to Bernhard Sadlowski
1400 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
1401 and helping me resolve this and other minor libc5 compiling
1404 2. Fix a problem when dumping a ext2fs with the 'filetype'
1405 feature enabled. Thanks to Patrick J. LoPresti
1406 <patl@cag.lcs.mit.edu> for reporting the bug and to
1407 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
1409 3. Made the nodump flag work on directories. A directory which
1410 has the nodump flag gets never dumped, regardless of its
1413 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
1414 which allows dump on an active ext3 filesystem. However, this
1415 is a "quick and dirty" patch which enables backup of an ext3
1416 filesystem through the ext2 compatibility (by ignoring the
1417 NEEDS_RECOVERY bit). The journal file is not recognized and
1418 it is dumped (it should not).
1420 5. Test the superblock compatibility flags when dumping, in order
1421 to be sure that we know how to deal with specific features.
1423 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
1424 ====================================================================
1426 1. Use lchown instead of chown, fixing a possible security problem
1427 when restoring symlinks (a malicious user could use this
1428 to deliberately corrupt the ownership of important system files).
1429 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
1430 this and providing the patch.
1432 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
1433 ====================================================================
1435 1. Put dump sources under CVS, added Id tags in all files so
1436 one can use 'ident' on binary files.
1438 2. Added the dump/restore version in the usage text so one can
1439 easily verify the version he is using.
1441 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
1442 a va_start/va_end problem on linux-ppc (always call va_start
1443 va_end in pairs each time we use a vararg function).
1445 4. Added again the DT_* constants because old libc does not
1446 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
1447 for submitting the bug report.
1449 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
1450 this should enable dumping big (huge) filesystems.
1452 6. Added the RSH environment variable in order to be able to
1453 use a rsh replacement like ssh when doing remote backups (and
1454 bypass the security limitations of rcmd). Now you can do remote
1455 backups without being root (or making dump setuid root).
1457 7. Modified again the way dumpdates works. For incremental dumps,
1458 we need to read dumpdates even if we are not using 'u' option.
1459 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
1462 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
1463 ===================================================================
1465 1. Removed the 'k' flag from the restore 'about' text if kerberos
1466 was not compiled in.
1468 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
1469 to them (fsetflags takes a char*, setflags an open fd!).
1471 3. (f)setflags is called only if the flags aren't empty. If the
1472 file is a special file, a warning is printed, because changing
1473 flags implies opening the device. Normally, a special file
1474 should not have any flag... (Debian bug #29775, patch provided
1475 by Abhijit Dasgupta <abhijit@ans.net>).
1477 4. Made possible to dump a file system not mentioned in /etc/fstab.
1478 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
1480 5. Changed the default behaviour to not create dumpdates
1481 unless 'u' option is specified. Removed the old "debian-patch"
1482 which provided the same thing. (Debian bug #38136, #33818).
1484 6. Removed all those dump*announce, since they were getting old...
1486 7. Added warning messages if dumpdates does not exist and
1487 when an inferior level dump does not exist (except for a level 0
1490 8. Debugged the glob calls in interactive mode: restore used a
1491 dirent struct which was different from the /usr/include/dirent.h
1492 one (this used to work, is it a glibc2 change?), so none of the
1493 compat glob (which used /usr/include/dirent.h) or the system glob
1494 worked. Restore use now the system dirent (and the system
1495 DT_* constants), which are compatible with BSD ones.
1497 9. Added a configure flag (--with-dumpdatespath) to specify
1498 the location of dumpdates file. By default, it is
1501 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1502 included the current date/version in man pages.
1504 11. Corrected the estimation of remaining time when
1505 the operator doesn't change the tapes quickly enough. This
1506 was an old bug, I thought I corrected it, and discovered
1507 that in fact it was corrected in two different places, so
1508 the results canceled each other...
1510 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1511 ===================================================================
1513 1. Integrated multiple patches from RedHat, Debian and SuSE:
1515 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1517 - several fixes in the man pages.
1518 - update the default tape device to /dev/st0.
1519 - many updates for Linux Alpha (byte ordering, size_t etc).
1521 - use environment variable for TMPDIR (instead of /tmp).
1522 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1523 - workaround egcs bug (RedHat bugs #4281 and #2989).
1524 - wire $(OPT) throughout Makefile's.
1526 2. Upgrade the dump revision to 1, making possible to dump filesystems
1527 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1529 3. Fix some compile warnings, prototype all functions.
1531 4. Use glibc err/glob instead of internal compatibility
1532 routines (only if available).
1534 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1535 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1537 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1538 ======================================================================
1540 1. Integrated the changes from FreeBSD-3.1-RELEASE
1541 (mostly bug fixes, buffer overruns, dump has now an "automatic
1542 tape length calculation" flag, dump/restore can use kerberos now
1543 (this is NOT tested), use environment variables for TAPE and
1546 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1548 3. Corrected a bug in restore when using 'C' option with multi-volumes
1549 tapes (files splited accros two tapes give "size changed" errors
1552 4. Corrected the long standing bug when dumping multiple tapes.
1553 This works for me, needs further testing.
1555 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1556 ====================================================================
1558 1. Dump now runs correctly on kernels 2.1.x
1559 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1561 Changes between versions 0.4b2 and 0.4b3
1562 ========================================
1564 1. Use realpath() if available
1566 2. Report statistics
1568 Changes between versions 0.4b1 and 0.4b2
1569 ========================================
1571 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1574 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1576 3. Fixed configure.in to correctly handle --enable-rmt
1578 Changes between versions 0.3 and 0.4b1
1579 ======================================
1581 1. Integrated the changes from 4.4BSD-Lite2
1583 2. Integrated the patches from Debian and Red Hat
1585 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1587 4. Changed dump to use the Ext2fs library to get block addresses. This
1588 should solve the endianness problem on SparcLinux.
1590 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1591 distribution's one) to use autoconf
1593 6. Fixed a few minor bugs
1595 Changes between versions 0.2e and 0.2f
1596 ======================================
1598 1. Added the creation of named pipes (fifos) in restore.
1600 2. Added the -N flag in the restore manual page.
1602 3. Added the file kernel.patch which contains the llseek() optimization
1603 patch for 1.2.x kernels.
1605 4. Fixed a bug in the restoration of symbolic links: owner and group were
1608 5. Integrated some changes from FreeBSD 2.2.
1610 6. Added a call to ftruncate() after restoring each file to restore
1611 correctly files ending by a hole.
1613 Changes between versions 0.2d and 0.2e
1614 ======================================
1616 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1617 with owner = root (0) and group = root/wheel/whatever (0).
1619 Changes between versions 0.2c and 0.2d
1620 ======================================
1622 1. Dump is now able to backup 2GB+ filesystems.
1624 2. Dump and restore can now be linked as static binaries.
1626 Changes between versions 0.2b and 0.2c
1627 ======================================
1629 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1631 Changes between versions 0.2a and 0.2b
1632 ======================================
1634 1. Really fixed the bug that I should have corrected in 0.2a.
1636 2. Enabled optimization again.
1638 Changes between versions 0.2 and 0.2a
1639 =====================================
1641 1. Disabled the optimization during compilation.
1643 Changes between versions 0.1 and 0.2
1644 ====================================
1646 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1647 the fs_type field (actually, I modified the file fstab.c to make it
1648 use the mntent functions).
1650 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1651 made by dump should be readable by the BSD restore and backups made
1652 by the BSD dump should be readable by restore. Unfortunately, this
1653 means that the dump format has changed between version 0.1 and version
1656 3. Dump is now able to backup a subtree, it is no longer limited to whole
1657 filesystems like the BSD version.
1659 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1662 Changes between versions 0.0 and 0.1
1663 ====================================
1665 1. Now create links rdump and rrestore during the `make install' step.
1667 2. Linux port specific bugs added to the manual pages
1669 3. Incorrect estimation of the number of tapes blocks fixed when doing
1670 incremental backups.
1672 4. Better ls-like format in restore in interactive mode.