1 $Id: CHANGES,v 1.286 2007/02/22 20:16:23 stelian Exp $
3 Changes between versions 0.4b41 and 0.4b42 (released xxxxxxxxxxxxxxx)
4 =====================================================================
6 1. Fix printout of the 'Connection to' message in dump/restore.
7 Thanks to Dale Wiles <dwiles@users.sourceforge.net> for the
8 bug report and its fix.
10 2. Fix dump -w/-W output for filesystems having the last backup
11 level equal to 0. Thanks to Pascal Bouchareine
12 <kalou@users.sourceforge.net> for reporting the bug.
14 3. Enable restore to handle restoring onto a different SELinux
15 policy from the dump. Thanks to Tony Nelson
16 <tony_nelson@users.sourceforge.net> for the patch.
17 (see http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=189845
20 4. Fix EA set failures when restoring immutable files. Thanks
21 to Andrew Kroeger for the patch.
23 Changes between versions 0.4b40 and 0.4b41 (released January 2, 2006)
24 =====================================================================
26 1. Fix restore of dumped Access Control Lists. The previous
27 code for EA works fine, but ACLs needed conversion from the
28 ext2/3 disk format to posix_acl format before restoring.
30 2. Fix some issues with restoration of EA on big endian
33 3. Fix restore when the symtab is over 2GB in size.
35 4. Made the directory hash indexing an optional feature,
36 accessible by the '-H' option of restore, and disable it
39 5. Fixed dump to not include extended attributes information
40 in the toc (archive) file which confused restore -t.
42 Changes between versions 0.4b39 and 0.4b40 (released May 2, 2005)
43 =================================================================
45 1. Changed restore to emit warnings (instead of emitting a fatal
46 error) if a file (or a directory) is unavailable for a
47 comparision (if the user doesn't have the necessary permissions
48 to access it for example). Thanks to Kenneth Porter
49 <shiva@sewingwitch.com> for the bug report.
51 2. Re-done the 'do not save directory entries to non-dumped inodes
52 (excluded from dump)' feature. The previous implementation
53 worked well for excluded directories but not for regular files.
54 Thanks to Kenneth Porter <shiva@sewingwitch.com> for the bug
57 3. Fixed a bug in dump where the tape size was miscalculated when
58 the user used -d/-s to specify the tape characteristics. Thanks
59 to Philip Goisman <goisman@physics.arizona.edu> for reporting
62 4. Fixed another bug introduced in restore with the hashtree
63 implementation. This one caused restore to stop saying
64 "removenode: non-empty directory" in some cases.
66 5. Added support for dumping and restoring ext2/3 extended
67 attributes (EA), like the access control lists (ACL) or
68 the security labels used by SELinux.
70 Changes between versions 0.4b38 and 0.4b39 (released January 21, 2005)
71 ======================================================================
73 1. The newly added dump_on_cd_3 example was buggy, replace it
74 with an updated version from Andrew Basterfield
75 <bob@cemetery.homeunix.org>.
77 2. Made restore to chdir() back into the initial directory when
78 dumping core while aborting a comparision operation. The
79 previous behaviour was to write the corefile at the root of
80 the directory being compared, which could very well be
81 read only and preventing the corefile generation. Thanks
82 to Kenneth Porter <shiva@sewingwitch.com> for the bug report.
84 3. Silenced the failure to call fgetflags() when comparing an
85 entry which has no ext2 attributes (as in lsattr()).
87 4. Fix a brown paper bug in restore -C which broke restore and
88 caused modifications on the filesystem being compared
89 (directories containing a file with the same name as the
90 directory get renamed to RSTTMP...). Thanks to Kenneth Porter
91 <shiva@sewingwitch.com> for finding the bug and helping me
94 5. Made restore -C force the -N flag (no writing allowed on
95 the disk) in order to prevent more bugs like the above one.
97 Changes between versions 0.4b37 and 0.4b38 (released January 7, 2005)
98 =====================================================================
100 1. Fix a couple of troff syntax bugs in the man pages.
101 Thanks to Eric Raymond <esr@thyrsus.com> for the patch.
103 2. Made restore use either libncurses or libtermcap, depending
104 on which one is available at configure time.
106 3. Fixed restore negative size display bug when comparing a
107 dump containing files over 2GB. Thanks to Steve Bonds
108 <sbonds@users.sourceforge.net> for the bug report.
110 4. Do not save directory entries to non-dumped inodes
111 (excluded from dump). This will eliminate the 'missing
112 file' warnings when doing 'restore -C'.
114 5. Fix dump crash when backuping a huge (2TB) filesystem,
115 due to a bogus calculation on the inode map size.
116 Thanks to Kevin B. Haines <K.B.Haines@rl.ac.uk> for
117 submitting the bug and testing the fix.
119 6. Fix a problem in restore where the final \0 in the symbolic
120 link names could have been lost, generating corrupt filenames.
121 Thanks to Kyle Wilson <kyle.wilson@amd.com> for reporting the
124 7. Implemented a hash list for the directory names in restore.
125 The linear list used before caused problems in interactive
126 restores when dealing with directories having thousands of
127 entries. Thanks to Brian Ristuccia <bristuccia@starentnetworks.com>
128 for reporting the bug.
130 8. Improved restore -C, this time including the directory
131 attributes into the comparision.
133 9. Made restore understand tapes containing EA/ACLs (which will
134 be dumped by the next version of dump). In this version
135 extended attributes on the tape are ignored, for full EA/ACL
136 support wait for the next version or try the experimental EA
139 Changes between versions 0.4b36 and 0.4b37 (released July 7, 2004)
140 ==================================================================
142 1. Added the --enable-staticz configure option which enables
143 dump and restore to be linked with static versions of
144 libz and libbz2 (and dynamic versions of all other libraries).
145 This will make Debian users happy, because libz and libbz2
146 were the only needed libraries living in /usr, all the
147 others live in /lib. In case of system emergency, it is
148 better not to have to rely on an extra filesystem.
149 Thanks to Bdale Garbee <bdale@gag.com> for the suggestion.
151 2. Fix compilation on (at least the Linux Debian port to) AMD64.
152 (<ext2fs/ext2_types.h> defines some types (__s64 and __u64)
153 that are also defined by <linux/types.h> (<asm/types.h>) and
156 3. Make dump's reading of the dumpdates file a bit more robust,
157 preventing dump from crashing when the dumpdates file has
158 been modified by hand.
160 4. Fixed some offset calculations in dump code which could
161 lead to "bread lseek errors" on large filesystems. Thanks
162 to Bruce Lowekamp <lowekamp@users.sourceforge.net> for
163 reporting this bug and debugging the issue.
165 5. Made dump use the blkid library when searching for devices
166 by label or uuid instead of dump's own routines.
168 6. Corrected a bug in dump where a wrong LABEL=... line in
169 /etc/fstab could prevent dump from dumping unrelated
170 filesystems. Thanks to Bruce Lowekamp
171 <lowekamp@users.sourceforge.net> for reporting the bug.
173 Changes between versions 0.4b35 and 0.4b36 (released April 21, 2004)
174 ====================================================================
176 1. Fixed dump compilation with old gcc versions. Thanks to
177 Mike Castle <dalgoda@users.sourceforge.net> for the patch.
179 2. Fixed some warnings (howmany, roundup, powerof2 redefined)
180 when compiling against a recent glibc version.
182 3. Fixed a bug in restore preventing the read of a dump tape
183 written with Solaris 7 ufsdump. Thanks to Patrick Higgins
184 <phiggins@transzap.com> for reporting the bug and providing
187 4. Changed dump to enable the creation of volumes of different
188 sizes in a single run (make -B accept a list of values).
189 Patch contributed by Florian Zumbiehl <florz@gmx.de>.
191 5. Use the glibc provided minor() and major() macros instead
192 of our own bitmask implementation. This should be safe for
193 when the major/minor namespace will migrate to 32 bits.
194 Thanks to Zhang Jun <zhangjun@nanjing-fnst.com> for reporting
197 6. Made explicit in the dump man page that dump will not create
198 a remote file, it will only write to an already existing one.
200 7. Another try at making size estimates better again.
202 8. Put back the inconditional running of the end-of-tape
203 script which was changed in 0.434 to be run only when -M
204 or multiple -f were NOT specified. Some users rely on this
205 feature even when it is combined with -M/-f.
207 9. Fixed restore when restoring huge backups (where rstdir...
208 temporary files are over 2GB). Thanks to
209 Raphael Jaffey <rjaffey@artic.edu> for reporting this,
210 debugging the issue and making the patch.
212 10. Made restore understand FreeBSD UFS2 tapes. Thanks to
213 David <vrtin@users.sourceforge.net> for submitting the bug
214 and providing a test case.
216 11. Made dump work with an arbitrary integer level (no more 10
217 levels only limitation). Thanks to Matthew
218 <msvincen@midway.uchicago.edu> for the patch.
220 Changes between versions 0.4b34 and 0.4b35 (released December 21, 2003)
221 =======================================================================
223 1. Added a note in the dump man page saying that the default
224 blocksize can be 32 if -d is used with a high density argument.
225 Thanks to Antonios Christofides <A.Christofides@itia.ntua.gr>
228 2. Fixed configure to correctly understand CPPFLAGS, CFLAGS,
229 LDFLAGS environment variables. Thanks to Arcady Genkin
230 <antipode@users.sourceforge.net> for reporting the bug.
232 3. Made -e/-E options of dump accept an unlimited number of inodes
233 to be excluded, rather than a hardcoded maximum. Thanks to
234 Dietrich Rothe <d-rothe@users.sourceforge.net> for the patch.
236 4. Updated the autoconf system to 2.50. Forced the -D_BSD_SOURCE
237 and -D_USE_BSD_SIGNAL defines in configure in order to solve
238 64bit build problems because quad_t is redifined with a
239 different signature. Thanks to Mike Harris <mharris@redhat.com>
240 for reporting this bug.
242 5. Made restore build on Solaris, making possible to
243 restore Linux's "enhanced" tapes. Thanks to Uwe Gohlke
244 <uwe@ugsoft.de> for the patch.
246 6. Made an extension in the dump tape format capable of saving
247 MacOSX specific inode extensions. Uwe Gohlke <uwe@ugsoft.de>
248 wrote the extension and contributed the restore code back
249 into this codebase. The same extension mechanism will be
250 used in the future to save ACLs...
252 7. Made rmt work correctly with regard to QFA and local/remote
253 files and tapes. The remote access will however work only
254 when the dump provided rmt version is used. If you want to
255 use another rmt server, please do not use the QFA feature.
256 Thanks to Uwe Gohlke <uwe@ugsoft.de> for the patch.
258 Changes between versions 0.4b33 and 0.4b34 (released April 18, 2003)
259 ====================================================================
261 1. Fixed the INSTALL file to reflect the actual install paths.
262 Thanks to David Ronis <ronis@ronispc.chem.mcgill.ca> for
265 2. Fixed the configure script to only check for headers presence
266 instead of trying to compile them. This should fix issues
267 with old build environments. Thanks to Kari Lempiainen
268 <kari@funky.pp.fi> for reporting the bug.
270 3. Fixed restore to correctly ignore sockets when comparing
271 a dump (as socket cannot be properly restored anyway). Thanks
272 to Gunther Reiszig <gunther@mit.edu> for reporting the bug.
274 4. Fixed restore to correctly access the archive file (-A argument)
275 even when using a remote tape. Thanks to Fabrice Bellet
276 <fabrice@bellet.info> for reporting the bug.
278 5. Fixed (again) handling of long (largefile) seeks in rmt.
279 Thanks to Fabrice Bellet <fabrice@bellet.info> for reporting
282 6. Fixed restore corner case when dealing with large block sizes
283 dump is able to write now (-b 1024). Thanks to Fabrice Bellet
284 <fabrice@bellet.info> for reporting the bug.
286 7. Fixed a bug preventing dump to access a filesystem having
287 a label exactly 16 bytes in length. Thanks to <minduni@ti-edu.ch>
288 for reporting the bug.
290 8. Made dump store dump dates using explicit timezones, fixing a
291 problem with incremental dumps if the timezone is modified
292 between the dumps. Thanks to Philippe Troin <phil@fifi.org> for
293 the bug report and the patch.
295 9. Fixed a bug encountered when dumping individual files (not full
296 filesystems or directories) and dangling symbolic links happen
297 to be in the list of files. For as far as dump is concerned,
298 dangling symbolic links are allowed, and are dumped as is.
299 Thanks to Jin-su Ahn <jsahn@ee.snu.ac.kr> for reporting the
300 bug and providing the fix.
302 10. Fixed open and creation modes and permissions for QFA and
303 table-of-contents files in dump and restore. Thanks to
304 Philippe Troin <phil@fifi.org> for the patch.
306 11. Fixed the archive file descriptor handling enabling it to be 0.
307 This can happen in some cases when shell redirections are used.
308 Thanks to Philippe Troin <phil@fifi.org> for the patch.
310 12. Delayed the opening of archive file until after suid had been
311 dropped (fixing a possible security issue if dump is suid).
312 Thanks to Philippe Troin <phil@fifi.org> for the patch.
314 13. Fixed the 'S' command handling in the rmt client part.
315 Thanks to Philippe Troin <phil@fifi.org> for the patch.
317 14. Modified the end-of-tape script handling to print out statistics
318 (and stop the timer) before launching the eot script. Also, the eot
319 script does not get run anymore when using -M (which makes sense) or
320 when multiple tapes are listed on the command line
321 (-f tape0,tape1,tapen) (which also makes sense).
322 Thanks to Philippe Troin <phil@fifi.org> for the patch.
324 15. Relicensed dump/restore under the 'revised' BSD license, as per
325 ftp://ftp.cs.berkeley.edu/ucb/4bsd/README.Impt.License.Change.
327 16. Added LZO compression to dump. This new compression method has
328 the advantage of being super fast, thus not killing tape streaming
329 on slow machines. Thanks to Helmut Jarausch
330 <jarausch@igpm.rwth-aachen.de> for the patch and to
331 Markus Oberhumer <markus@oberhumer.com> for giving special permission
332 to include his miniLZO project (GPL licensed) in dump/restore.
334 17. Some small buffer overruns fixes in rmt. Thanks to Antonomasia
335 <ant@notatla.demon.co.uk> for reporting the bugs.
337 18. Added a special rmt version which can do encryption when writing
338 to tape. Read examples/encrypted_rmt/README for details on
339 how to enable and configure it. Thanks to Ken Lalonde
340 <ken@globalremit.com> for the patch.
342 19. Made dump work with 2.5 kernel end of tape early warning semantics.
343 Thanks to Kai Makisara <Kai.Makisara@kolumbus.fi> for the patch.
345 20. Fixed a bug which caused dump -w|-W not to work anymore, because
346 the fs_freq and fs_passno fields in /etc/mtab are always set
347 to 0 0. Thanks to Trent Piepho <xyzzy@speakeasy.org> for
350 Changes between versions 0.4b32 and 0.4b33 (released February 10, 2003)
351 =======================================================================
353 1. Added a note in the restore man page clarifying the question
354 'set the permissions on the current directory ?' asked by
355 restore at the end of treatment in -i and -x modes.
357 2. Fixed the endianess issues when reading compressed tapes.
358 Thanks to Dark Force <daq4th@users.sourceforge.net> for
359 reporting this bug and providing test cases.
361 3. Fixed the "ACL won't be dumped" warning message (which showed
362 an extra, unrelated error message). Thanks to Dragan Krnic
363 <dkrnic@lycos.com> for reporting this bug.
365 4. Made dump look first into /etc/mtab, then into /etc/fstab
366 when searching for filesystem paths. Also fixed some problems
367 caused by binding mounts or multiple block device mounts.
368 Thanks to Matus Uhlar <uhlar@fantomas.sk>, Elliott Mitchell
369 <ehem@m5p.com>, Greg Edwards <gedwards@users.sourceforge.net>,
370 Brian Hoy <brian.hoy@opus.co.nz>. (fixes Debian bugs #147086
371 and #160305, Sourceforge bugs #618699 and #679832).
373 5. Made dump's -I option accept the value '0' meaning all the
374 read errors will be ignored. This can be useful when running
375 dump from unattended sessions (like cron jobs). Thanks to
376 John I Wang <jiwang@users.sourceforge.net> for the suggestion.
378 6. Fixed the output of dump to indicate 'blocks' instead of
379 'tape blocks' in the various messages (blocks are always
380 1 Kilobyte, tape blocks are 1 BK * '-b' argument), and
381 made it clearly print the current blocksize at the start of
382 a dump. Thanks to Michal Szymanski <msz@astrouw.edu.pl> for
385 7. Made rmt understand long (largefiles) seeks.
387 8. Fixed build with very old versions of libext2fs, where
388 EXT2_FT_* constants were undefined.
390 9. Made dump accept the dumpdates path on the command line
391 (-D file option) instead of using only the hardcoded one.
392 Thanks to Piete Brooks <pb22@users.sourceforge.net> for the
395 10. Enabled rmt, LFS, readline, QFA options by default in
396 ./configure. Updated the configure process (new versions
397 of config.guess, config.sub etc).
399 Changes between versions 0.4b31 and 0.4b32 (released November 15, 2002)
400 =======================================================================
402 1. Changed dump to use fcntl(F_SETLK) style locking instead
403 of flock() when locking the dumpdates file. With the old
404 locking scheme, a local user having read rights on the
405 dumpdates file could be able to do a Denial of Service attack
406 on dump. In order to lock the dumpdates file with the new
407 scheme, the user would need to have write access on the file.
408 Thanks to Richard Johnson <Richard.Johnson3@ey.com> for
409 reporting the bug (originally a bugtraq post).
411 2. Fixed interactive 'ls' which caused spurious errors warnings
412 about 'undefined filetypes' detected. Thanks to Jorgen Ostling
413 <jorgen_ostling@users.sourceforge.net> for reporting this
416 3. Fixed dump's estimate when dealing with sparse inodes.
418 4. Modified dump to allow setting a blocksize bigger than 32kB
419 (raised the limit to 1024kB), because newer hardware needs
420 this for proper operation. Thanks to Dirk Traenapp
421 <dtraenapp@users.sourceforge.net> for reporting this.
423 5. Fixed a bug causing Dump to stop and report an error if an
424 inode number in the exclude file was followed by some amount
425 of whitespace. Thanks to Jeffrey Sofferin
426 <sofferin@users.sourceforge.net> for reporting this bug.
428 6. Fixed a bug which caused restore, in some particular cases,
429 to ask some 'scary' questions and leave a bunch of RSTTMP
430 directories behind when restoring incremental tapes. Thanks
431 to Philippe Troin <phil@fifi.org> for reporting this bug and
432 providing the test cases.
434 7. Changed the wording when inodes are excluded from dump:
435 replaced 'Added inode 7 to exclude list' with
436 'Excluding inode 7 (resize inode) from dump', as suggested
437 by Elliott Mitchell <ehem@m5p.com> in a Debian bug report.
439 Changes between versions 0.4b30 and 0.4b31 (released July 30, 2002)
440 ===================================================================
442 1. Fixed rmt open flags transmission (GNU's symbolic syntax over
443 rmt) which I broke in 0.4b29. Thanks to Eros Albertazzi
444 <eros@lamel.bo.cnr.it> for reporting the bug.
446 Changes between versions 0.4b29 and 0.4b30 (released July 25, 2002)
447 ===================================================================
449 1. Made dump print out the ext2 logical block number in case of a read
450 error, which can be given as an argument to debugfs' ncheck command,
451 in order to find out the affected inode. Added note about this
452 usage in dump's man page.
454 2. Fixed a problem in restore when reading tapes written on big
455 endian machines with very old versions of dump. The patch was
456 contributed by George Helffrich <george@geology.bristol.ac.uk>.
458 3. Fixed the tape length calculation when using large tapes
459 and compression. Thanks to Georg Lippold
460 <g_lippold@sourceforge.net> for reporting the bug.
462 4. Added a new set of examples in dump_on_cd_2 directory, based
463 on dump_on_cd examples but somewhat enhanced, supporting DVD
464 media, and localized in english and german. Thanks to
465 Georg Lippold <g_lippold@sourceforge.net> for the new scripts.
467 5. Made dump save 32 bit UID/GID. Internally, this was achieved
468 by switching from the old BSD inode format to the new BSD
469 inode format, which means that the tape format was changed.
470 However, since all restore versions out there should
471 transparently support both inode formats, the change should
472 have no side effects. Thanks to John Yu <jky@cs.bu.edu> for
475 6. Fixed a lot of warnings in the code shown when compiling
476 with 'gcc -W'. Thanks to Matthias Andree
477 <matthias.andree@stud.uni-dortmund.de> for reporting this.
479 7. Fixed a small markup bug in the dump man page. Thanks to
480 Eric S. Raymond <esr@minx.thyrsus.com> for submitting the
483 8. Rewrote entirely the man pages using the tmac.an macro
484 package (Linux man page format) instead of the original BSD
485 format. They should be now cleaner and easier to modify.
487 Changes between versions 0.4b28 and 0.4b29 (released June 8, 2002)
488 ==================================================================
490 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
491 non Linux clients were misinterpreted. The description of the problem
492 (incompatible numbering in Linux mtio opcodes) is documented at
493 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
494 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
495 providing an excellent, cross-platform replacement for rmt in his
498 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
499 which caused restore to fail when the inode maps were not entirely
500 contained on the first volume (when using really small volumes or
501 when dumping a huge number of inodes).
503 3. Fixed a problem in dump, where files in subdirectories of directories
504 that have the nodump flag set (or being excluded with -e/-E) were
505 sometimes incorrectly dumped (instead of being excluded). The original
506 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
507 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
508 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
509 help testing the patch.
511 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
512 which makes one able to pipe the output of dump, by the net, to
513 a remote CD-burner server.
515 5. Made dump use O_CREAT|O_TRUNC both locally and remotely (over rmt),
516 and use GNU's symbolic syntax over rmt instead of numerical values
517 to assure multiple platform compatibility.
519 6. Documented the -d option in restore.
521 7. Added a -v (verbose) mode to dump. For now it just prints the number
522 of the inode being dumped, but this could evolve in future versions
523 to include interesting debugging output.
525 8. Added a -o flag to restore, which automatically restores the current
526 directory permissions in -i or -x mode, without asking the operator
527 whether to do so. Patch submitted by Tony Clayton <tonyc@e-smith.com>
528 and Peter Samuel <peters@e-smith.com>.
530 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
531 ====================================================================
533 1. Fixed a bug in the QFA feature which could have caused
534 a bad tape positionning on restore (causing extra delays in
535 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
537 2. Added a small note in the dump man page specifying that
538 there should be no white space between the option letter and
539 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
541 3. Made restore work with older versions of the readline library,
542 by checking in configure for several extended symbols. Restore
543 can now be compiled with a readline as old as the 2.0 release
544 (though it may be a good idea to upgrade it to a more recent
545 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
546 reporting the build failures.
548 4. Fixed a performance problem with the QFA file creation in
549 dump, which made unnecessary seeks on the tape slowing down
550 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
551 reporting this issue.
553 5. Removed the inclusion of some kernel headers in the dump
554 source, which prevented the compile in some kernel/glibc
555 headers/architecture combination. Thanks to Bdale Garbee
556 <bdale@gag.com> for reporting the bug.
558 6. Added the appropriate error message when dump fails to
559 open the output file for writing. Thanks to Amith Varghese
560 <amithv@yahoo.com> for reporting this bug.
562 7. Made restore able to understand large Solaris ufsdump tapes
563 (containing inodes bigger than 4194304). Sun have introduced
564 an "extension" to the dump tape format when dealing with
565 those inodes, which was uncorrectly handled by Linux restore.
566 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
567 providing a test case.
569 8. Added the -m parameter to dump which optimises the output for
570 inodes having been changed but not modified since the last dump
571 ('changed' and 'modified' have the meaning defined in stat(2)).
572 For those inodes, dump will save only the metadata, instead of
573 saving the entire inode contents. Inodes which are either
574 directories or have been modified since the last dump are saved
575 in a regular way. Uses of this flag must be consistent, meaning
576 that either every dump in an incremental dump set have the flag,
579 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
580 =======================================================================
582 1. Fixed behaviour of dump when exceeding resource limits
585 2. Added the -L flag to restore to allow the user to specify a
586 maximal allowed number of miscompares when using restore
587 with the -C option to check the backup.
589 3. Detailed the manual entry for the -N option of restore.
591 4. Added the -a flag to restore to make able doing unattended
592 restores in -i or -x mode (automatically walks through the
593 multiple dump volumes).
595 5. Extended the QFA mode to work with local files and/or
596 remote tapes and files. This way, restore can know in advance
597 the tape number and the offset for the inodes to extract and
598 can minimize the extraction time by seeking directly to the
601 6. Added the -A <archive> option to both dump and restore,
602 which makes dump to archive a dump table-of-contents in
603 the specified file to be used by restore to determine
604 whether a file is in the dump file that is being restored.
605 (the archive file syntax is also compatible with the
606 Solaris ufsdump generated one).
608 7. Small fix in restore making it able to read some (broken ?)
609 Solaris ufsdump tapes.
611 8. Fixed dump to correctly recognise the root filesystem when using
612 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
613 for reporting this bug.
615 9. Added the -P <file> option to restore to create a
616 Quick File Access file from an already made dump. Patch
617 contributed by Uwe Gohlke <uwe@ugsoft.de>.
619 10. Made restore compile and run on Solaris, making it a
620 possible replacement for the standard ufsrestore. Port was
621 contributed by Uwe Gohlke <uwe@ugsoft.de>.
623 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
624 =====================================================================
626 1. Added a set of backup scripts from Eugenio Diaz
627 <getnito@yahoo.com> in the examples section. It features
628 automatic (cron based) full and incremental dumping of
629 several filesystems on a separate filesystem.
631 2. Fixed a off-by-one miscalculation which disabled dumping
632 a one letter subdirectory. Thanks to Chris Siebenmann
633 <cks@utcc.utoronto.ca> for reporting the bug.
635 3. Fixed several restore bugs occuring when trying to
636 restore some missing files on the tape. Thanks to Chris
637 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
639 4. Fixed --with-ldopts configure argument passing, installing from
640 a separate object directory, makefile cleanups contributed
641 by <splite@purdue.edu>.
643 5. Fix a bug which could caused, in some conditions, the highest
644 number inode of a filesystem, to not be dumped. Many thanks
645 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
648 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
649 =======================================================================
651 1. Added a mini howto from Patrick Walsh in the examples
654 2. Minor man pages syntax corrections. Thanks to
655 Chris Danis <screechco@home.com> for reporting the bugs.
657 3. Added a script from David B. Peterson <dave@toppledwagon.com>
658 to the examples section. It features dumping several
659 filesystems to a remote tape drive upon ssh.
661 4. Added a patch provided by Richard Jones <rich@annexia.org>
662 which allows BRADEMAX (number of read errors tolerated by
663 dump) to be adjusted using the -I option.
665 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
666 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
669 6. Add the -l option to restore to specify if, when doing a
670 remote restore, the file used to access the backup is a
671 regular file (the defaults being a tape drive). Restore needs
672 to know this information when reading a remote compressed
673 dump. Previously, this information was autodetected, but
674 the autodetection code fails (with ioctl: Inappropriate ioctl
675 for device) when using a non Linux remote box. Thanks to
676 many users and especially to Eros Albertazzi
677 <eros@lamel.bo.cnr.it> for reporting this.
679 7. Found a workaroung for the dump deadlock problem (3 childs
680 stuck in pause(), father in read()). The workaround seems
681 to work for me and several beta-testers. If it doesn't work
682 for you, please report back.
684 8. Updated the RPM spec file (BuildPrereq, URL etc).
686 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
687 ========================================================================
689 1. Fixed the permissions of a newly created QFA file by dump.
691 2. Cleaned up the source of dump (the external variables
692 definition was a complete mess, making possible to have
695 3. Fixed restore to use the full tape volume path when doing
696 a compare (since it changes the working directory to the
697 filesystem being compared in the process).
699 4. Added the -q option to dump which makes dump abort
700 whenever it needs operator attention. It should be
701 easier to use dump in scripts with this option.
703 5. Detect the use of incompatible options to dump and
704 refuse them (like -a and -B options together).
706 6. Added bzip2 compression to dump/restore (use option -j level
707 to select it). Note that you will need a restore version
708 >= 0.4b24 in order to restore a bzip2 compressed dump.
709 The same warning as for the zlib compression applies:
710 the tape format of a bzip2 dump is not compatible with the
711 original BSD tape format.
713 7. Fixed a overflow problem in dump corrupting the dump when
714 very large files were encountered. Thanks to Vladimir Nadvornik
717 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
718 kernel buffer/page cache before starting the dump, helping
719 a bit those who use dump on mounted filesystems. Thanks to
720 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
723 9. Updated the RPM spec file following the RedHat changes
724 (dynamically linked binaries now in /usr/sbin etc).
726 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
727 which enables restore to recognise multi volume compressed dumps
728 done on CD. Included his scripts for dump (which pipe the dump
729 output directly into cdrecord) and restore. There is now
730 possible to to backups to CD on the fly!
732 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
733 ===================================================================
735 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
736 Marcus Meissner (Caldera International Security Dept.).
738 2. Implement the Sun rmt extended protocol. Patch submitted
739 by Ian Gordon <iangordon@users.sourceforge.net>.
741 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
742 the linux kernel header. This ensures that dump always has the
743 latest version of this file. Patch submitted by
744 Andreas Dilger <adilger@turbolinux.com>.
746 4. Report any filesystem present in either /etc/fstab with a
747 positive passno or /etc/dumpdates in dump -w output.
748 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
750 5. Fixed the looping problem in dump introduced in the
753 6. Changed the -B option of dump to limit the size of
754 _compressed_ output per volume if compression is on.
755 Patch contributed by Helmut Jarausch
756 <jarausch@igpm.rwth-aachen.de>. Note however that, since
757 it is impossible to predict the size of the compressed
758 data before doing the compression, the -B limit is a bit
761 7. Fixed a bug in reading the operator typed file/tape path for
762 the next volume in restore.
764 8. Implemented a "-F script" option to restore which permits the
765 user to specify a script which will be launched at the
766 beginning of each tape, useful for automatic programming of
767 tape changers for example. See the restore man page for the
768 script parameters and return codes.
770 9. Small fix for the QFA routines provided by Uwe Gohlke
771 <uwe@ugsoft.de>, and some recommendations for QFA uses in
774 10. Fixed the multivolume restoring where making a mistake
775 to the 'Mount next tape' prompt caused several blocks to
778 11. Enhanced the -e option of dump to take as a parameter a
779 comma separated list of inode numbers.
781 12. Added the -E option to dump which specify a file containing
782 inode numbers to exclude from the dump.
784 13. Fixed the compressed multi-volume dump + restore.
786 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
787 ==================================================================
789 1. Made dump report the number of blocks dumped per volume.
790 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
792 2. Fix a bug in dump where waiting too much at the 'change volumes'
793 question caused the volume to be rewritten. Thanks to
794 Graham Reed <greed@users.sourceforge.net> for reporting the
795 bug and providing a patch.
797 3. Added a compression option to dump, contributed by Jerry
798 Peters <gapeters@worldnet.att.net>.
800 WARNING: the compression works only on files/pipes and
801 drives supporting variable block size.
803 WARNING2: tapes created using the compression feature are
804 incompatible with the standard dump tape format, and a
805 version of restore >= 0.4b22 will be needed for extraction.
807 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
808 architectures. Thanks to Paul Slootman <paul@debian.org> for
809 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
812 5. Many cleanups (CPP defines, const char warnings, check of
813 ext2fs COMPAT flags, time_t cleanups, added libext2 version
814 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
816 6. Made --prefix option work in configure. All the install path
817 are now based on the configure parameters.
819 7. Added the Quick File Access mode in dump/restore, contributed
820 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
821 a file tape position for each inode, and this file is used by
822 restore (if called with parameter Q and the filename)
823 to directly position the tape at the file restore is currently
824 working on. This saves hours when restoring single files from
825 large backups, saves the tapes and the drive's head. Use
826 --enable-qfa option of configure to compile in the QFA support.
828 8. Added the possibility to dump several files and directories
829 in a single invocation of dump. Thanks to Uwe Gohlke
830 <uwe@ugsoft.de> for implementing this option.
832 9. Fixed the dumping and restoring of files with holes
833 (files ending with holes didn't get dumped or restored
836 10. Fixed a socket descriptor leak in dump, which leaved opened
837 3 file descriptors per dump process (and there is one dump
840 11. Fixed dump large file system support, by explicit use of
841 open64/lseek64/etc functions (explicit use needed because
842 e2fsprogs libraries don't behave well when compiled with
843 FILE_OFFSET_BITS=64).
845 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
846 ======================================================================
848 1. Fixed some bugs in the dump -w|-W logic introduced by
849 the previous version. Thanks to Andreas Dilger
850 <adilger@turbolinux.com> for his help on this one.
852 2. Fixed again a compilation problem when using old e2fs
853 headers (filesystem label related). Thanks to many users
854 who reported this stupid error.
856 3. Fixed a build problem on old lib5 based systems dealing with
857 _PATH_MNTTAB being undefined. Thanks to John Adams
858 <johna@onevista.com> for reporting the bug.
860 4. Improved the error detection in restore when creating
861 the temporary files in TMPDIR. Now restore will corectly
862 report a 'no space left on device' error instead of
863 strange errors which could imply an archive corruption.
864 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
865 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
867 5. Added the throughput information in dump's progression
868 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
871 6. Use libext2fs's inode scanning routines, which are particularly
872 robust when dealing with errors (especially when having some
873 bad blocks on the filesystem). Many thanks to Colin
874 <colin@horizon.com> for reporting the bug, investigating
875 the issues, finding a workaround, writing the patch and
876 fully testing it... (of course, if this patch will break
877 anything, he is to blame for :-)).
879 7. Made dump and restore LFS aware. Dump can dump a filesystem
880 containing large files, generate a large file on output and
881 restore can restore them. This feature is not enabled by
882 default, use --enable-largefile option of configure to enable
883 it (you need to have a LFS aware glibc though). Thanks to
884 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
885 and to Theodore T'so <tytso@valinux.com> for his always
888 8. Made dump ask upon a tape write error if it should rewrite
889 the current volume (assume this is a bad tape) or if it should
890 assume an end-of-tape condition (useful for tape drives which
891 behaves badly at the end of the tape). Thanks to Andreas
892 Dilger <adilger@turbolinux.com> for the suggestion.
894 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
895 =======================================================================
897 1. Fixed a small compilation problem due to a change
898 in the definintion of the struct sigaction in
899 glibc 2.0 / libc5. Thanks to Gunther Schlegel
900 <schlegel@riege.de> for reporting the bug and to
901 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
904 2. Modified the label and uuid analysis in order to be
905 self-contained (independant of kernel/libc headers). This
906 should fix the compile with older kernel/libc headers and
907 will preserve the functionality. Thanks to Bernhard Erdmann
908 <bernhard.erdmann@gmx.de> for reporting the bug.
910 3. The 'exclude inode' option, if applied to a directory
911 inode, excludes now correctly all the leaves of this
912 directory. Thanks to John R. Dennison
913 <gerdesas@users.sourceforge.net> for reporting the bug.
915 4. Fixed the '-e' option to disable the possibility
916 to exclude the root inode (which causes the dumps to
917 be unreadable by restore). Prevented array overflow
918 when multiple -e options are used.
920 5. Fixed dump to correctly interpret a filesystem argument
921 which is a mountpoint and it is not an absolute path
922 (as specified in the fstab). Thanks to Bernhard R. Erdmann
923 <be@berdmann.de> for reporting the bug.
925 6. Made dump able to backup files larger than 2 GB. Note that
926 dump still doesn't cope with files larger than 4 GB.
928 7. Restore the real uid and gid before invoking an external
929 RSH program (big hole when dump or restore is suid root!).
931 8. Get the values for the default tape device and the location
932 of fstab file from the system headers. Thanks to
933 Andreas Dilger <adilger@turbolinux.com> for the patch.
935 9. Made dump -w|-W report all recognized filesystems
936 present in either /etc/fstab or /etc/dumpdates, and present
937 the list in the same order as in fstab file. Thanks
938 to Andreas Dilger <adilger@turbolinux.com> for the patch.
940 10. Made dump's -a (automatic end of tape detection) the
941 default. Specifying one of -B, -c, -d or -s options will
942 override the EOM detection. Thanks to Andreas Dilger
943 <adilger@turbolinux.com> for the patch.
945 11. Save the ext2 filesystem label into the dump volume label.
946 Specifying a volume label on the command line (-L option)
947 will override this feature. Thanks to Andreas Dilger
948 <adilger@turbolinux.com> for the patch.
950 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
951 =====================================================================
953 1. Fixed the signal handling in dump (which I broke in 0.4b17)
954 which was causing several strange dump failures (dump
955 hanged or segmentation faults etc).
957 2. Specified the default blocksize in the dump man page.
959 3. Changed two info messages of restore to be written on stdout
960 instead of stderr in order to leave stderr only for errors
961 or warnings. Thanks to Stephen Atwell
962 <satwell@urbana.css.mot.com> for the suggestion.
964 4. Corrected an off by one calculation which prevented
965 dumping a directory having a 1 character name length.
966 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
967 for reporting the bug.
969 5. Reinforce dump in order to allow only level 0 and no
970 -u option when dumping a subdirectory, and document
971 this in the man page. Thanks to Bernhard Erdmann
972 <bernhard.erdmann@gmx.de> for reporting the bug.
974 6. Fixed a small harmless bug in restore which caused
975 in some conditions hard links to be restored several
976 times (therefore generation some warning messages).
977 Thanks to Kenneth Porter <shiva@well.com> for
980 7. Updated the RPM spec file to the latest RedHat version,
981 providing FHS packaging and other cosmetic changes.
982 You will need RPM version >= 3.0.5 in order to build the RPM.
984 8. Updated the configure script to check for libtermcap
985 before checking for libreadline, because we need this
986 library in order to compile the readline support.
988 9. Made dump understand the LABEL= and UUID= notation
989 both in /etc/fstab and on the command line. Note that
990 you will need the /proc filesystem in order to use
991 these notations. Thanks to Erik Troan <ewt@redhat.com>
992 for providing the patch.
994 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
995 ===================================================================
997 1. Fixed a potential buffer overflow in restore. Thanks
998 to Stan Bubrouski <satan@fastdial.net> for reporting
1001 2. Fixed a readline-related bug which prevented
1002 'cat DUMPFILE | restore -i -f -' from working. Thanks
1003 to Charles Karney <karney@users.sourceforge.net>
1006 3. Changed a few "panic" into "exit", causing restore to
1007 be more stable against some attacks (like the last one
1008 reported on Bugtraq, although the last version of restore
1009 was not vulnerable - just dumped core). Thanks to
1010 Andreas Hasenack <andreas@conectiva.com.br> for reporting
1013 4. Removed the suid-root bit on dump and restore in the
1014 default build (and generated RPMs). It should be safer
1015 now. Users who need the suid-root capabilities in order
1016 to make network backups should read first the man page
1017 of dump and enable the suid bit manually.
1019 5. Added -ltermcap to the compile parameters for restore
1020 when using readline support, in order to make the compile
1021 process work on some systems (SuSE ?). Thanks to
1022 Patrik Schindler <poc@pocnet.net> for reporting the bug.
1024 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
1025 ====================================================================
1027 1. The -F script is called now *only* at the end of a tape,
1028 not at the end of the dump. Thanks to Christian Haul
1029 <haul@informatik.tu-darmstadt.de> for the bug report.
1031 Normally, the device name and the current volume number
1032 are passed on the command line of the script. However,
1033 if you want the old style script (with no arguments
1034 passed on the command line), you can enable it in
1035 configure with the --enable-oldstylefscript.
1037 2. Use posix signal handling to preserve dump functionality
1038 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
1041 3. Made the exit code of restore in case of a 'C'ompare
1042 command reflect the result of the compare. An exit status
1043 of 0 means the dump archive is correct, 1 means tape errors,
1044 2 means that some files were modified. Thanks to Kenneth Porter
1045 <shiva@well.com> for the suggestion.
1047 4. Made (finally) quotes work in interactive restore.
1049 5. Small fixes in order to allow dump to compile with a
1050 really old e2fsprogs version. Thanks to Ian Zimmerman
1051 <itz@speakeasy.org> for the bug report.
1053 6. Add GNU readline capabilities to interactive restore.
1054 Use configure's --enable-readline flag to enable this feature.
1055 Thanks to Patrik Schindler <poc@pocnet.net> for the
1058 7. Do the compare on the fly in restore 'C' mode (this will
1059 allow not to exhaust the available /tmp space when
1060 ccmparing large files). Thanks to Kenneth Porter
1061 <shiva@well.com> for the suggestion.
1063 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
1064 ====================================================================
1066 1. Fixed some several duplicate 'DUMP: DUMP:' in the
1069 2. Corrected the estimation of blocks to dump. Note that
1070 this is still wrong for large directory inodes, due
1071 to the size of a BSD directory entry on the tape
1072 (estimation is based on the size of the ext2 directory,
1073 which is smaller because it doesn't contain the null
1074 character at the end).
1076 3. Made dump report the total number of MB written to
1077 tape at the end of run. Thanks to W. Reilly Cooley
1078 <wcooley@nakedape.cc> for the patch.
1080 4. Added the -X option to restore giving the possibility
1081 to read the names of the files to be extracted/listed
1082 from a text file (in addition of the command line).
1083 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
1086 5. Added the device name and the current volume number
1087 as arguments to the end of tape script (dump -F option).
1089 6. Made the multi-volume dump work again (after having
1090 broken it in 0.4b15).
1092 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
1093 ===================================================================
1095 1. Added a prompt command in interactive restore mode. Thanks
1096 to Andreas Dilger <adilger@home.com> for the patch.
1098 2. Fixed a buffer overflow problem in dump (caused by
1099 not checking the size of the filesystem parameter).
1100 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
1101 reporting this on Bugtraq (and to several dump users
1102 who forwarded me his mail).
1104 3. Added the '-F script' option to dump in order to
1105 launch a script at the end of each tape (to be used
1106 with a tape changer, or to notify the sysadmin by
1109 4. Fixed a bug in restore compare code caused by the changes
1112 5. Fixed the treatment of options using the old BSD syntax
1113 in both dump and restore.
1115 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
1116 =======================================================================
1118 1. Fixed a bug in dump which may have caused invalid deleted
1119 directories to be dumped out if they were deleted after the
1120 mapping pass. This could occure on active filesystem and lead
1121 to heap corruption (causing dump malfunction in many possible ways).
1122 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
1123 bug and submitting the patch.
1125 2. Fixed the handling of the filesystem parameter in dump. A
1126 '/mnt/disk' parameter caused the disk contents to be dumped,
1127 but a '/mnt/disk/' parameter caused the mountpoint directory
1128 to be dumped (generally an empty directory).
1130 3. Improved the output of dump in order to tell which directory
1131 it is currently dumping (when dumping a subtree).
1133 4. Added the '-e' exclude inode option to dump. Thanks to
1134 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
1136 5. Added a REPORTING-BUGS file in order to provide a guide
1137 on how to correctly report a bug in dump/restore.
1139 6. Fixed a restore problem when restoring a hard link to an inode
1140 having the immutable or append-only attribute set. Thanks to
1141 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
1143 7. Fixed a compatibility problem between dumps made on little
1144 endian machines (the format was correct) and big endian
1145 machines (the format was incorrect). This fix break the
1146 compatibility with the older dumps made on big endian
1147 machines (sparc, m86k, ppc etc). For the first time in
1148 linux dump's history (I believe), the dumps made by *BSD,
1149 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
1150 so interchangeable. Thanks to Rob Cermak
1151 <cermak@ahab.rutgers.edu> for submitting the bug and
1152 helping me test the fix.
1154 8. Fixed the way dump reports the remaining percent/time, if
1155 the number of blocks actually dumped exceeds the estimated
1156 number of blocks. Thanks to Jean-Paul van der Jagt
1157 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
1159 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
1160 ======================================================================
1162 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
1163 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
1165 2. Removed the requirement to build the RPM as root from the
1166 spec file. Thanks to Christian Weisgerber
1167 <naddy@mips.rhein-neckar.de> for submitting this.
1169 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
1170 causing dump to block when dumping really huge filesystems.
1171 Many thanks to Patrik Schindler <poc@pocnet.net> for
1172 helping me find this bug.
1174 4. Fixed the treatment for an interrupt signal when dump access
1175 the remote tape through RSH. Thanks to Christian Weisgerber
1176 <naddy@mips.rhein-neckar.de> for providing the patch.
1178 5. Fixed a bug which was causing dump/restore to display
1179 garbage characters instead of the remote host name.
1181 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
1182 =====================================================================
1184 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
1185 <kukuk@suse.de> for submitting the patch.
1187 2. Fix for the exit code when using the size estimate option of
1188 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
1189 submitting the patch.
1191 3. Handle EINTR in atomical reads/writes in dump, which was causing
1192 dump to fail on some systems. Thanks to Eric Jergensen
1193 <eric@dvns.com> for reporting the bug and submitting the patch.
1195 4. Handle more than 16 characters for the device names in dumpdates.
1196 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
1197 tracking down the problem and proposing the solution.
1199 5. Fixed a bug in dump which prevented the creation of the
1200 dumpdates file when doing a 0-level dump without already
1201 having a dumpdates file. Thanks to Patrik Schindler
1202 <poc@pocnet.net> for reporting the bug.
1204 6. Changed the way dump 'S' flag reports the size estimate
1205 from number of blocks into bytes (making it compatible
1206 with the Solaris version, and simplifying things for
1207 amanda users). Thanks to Jason L Tibbitts III
1208 <tibbs@math.uh.edu> for reporting the bug.
1210 7. Fixed a compatibility problem in linux/alpha dump tape format.
1211 Now the linux/alpha dump are (again) compatible with the
1212 other dump formats. But this breaks compatibility with
1213 older dumps made on alpha. Thanks to Mike Tibor
1214 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
1216 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
1217 ======================================================================
1219 1. Added a '--enable-kerberos' to configure.
1221 2. Added a 'S' option to dump which determines the amount of space
1222 that is needed to perform the dump without actually doing it, similar
1223 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
1224 <cermak@ahab.rutgers.edu>.
1226 3. Added a 'M' multi-volume option to dump and restore which enables
1227 dumping to multiple files (useful when dumping to an ext2
1228 partition to make several dump files in order to bypass the 2GB
1229 file size limitation). The 'f' argument is treated as a prefix and
1230 the output files will be named <prefix>001, <prefix>002 etc. With
1231 the 'M' flag, restore automatically selects the right file without
1232 asking to enter a new tape each time.
1234 4. Fixed a memory leak which was causing dump to grow very big
1235 (270MB when dumping a 10GB filesystem...). Thanks to Jason
1236 Fearon <jasonf@netrider.org.au> for reporting the bug.
1238 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
1239 ======================================================================
1241 1. Make configure test if the system glob routines support
1242 extended syntax (ALTDIRFUNC). If not, use the internal glob
1243 routines instead of system ones. Thanks to Bernhard Sadlowski
1244 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
1245 and helping me resolve this and other minor libc5 compiling
1248 2. Fix a problem when dumping a ext2fs with the 'filetype'
1249 feature enabled. Thanks to Patrick J. LoPresti
1250 <patl@cag.lcs.mit.edu> for reporting the bug and to
1251 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
1253 3. Made the nodump flag work on directories. A directory which
1254 has the nodump flag gets never dumped, regardless of its
1257 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
1258 which allows dump on an active ext3 filesystem. However, this
1259 is a "quick and dirty" patch which enables backup of an ext3
1260 filesystem through the ext2 compatibility (by ignoring the
1261 NEEDS_RECOVERY bit). The journal file is not recognized and
1262 it is dumped (it should not).
1264 5. Test the superblock compatibility flags when dumping, in order
1265 to be sure that we know how to deal with specific features.
1267 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
1268 ====================================================================
1270 1. Use lchown instead of chown, fixing a possible security problem
1271 when restoring symlinks (a malicious user could use this
1272 to deliberately corrupt the ownership of important system files).
1273 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
1274 this and providing the patch.
1276 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
1277 ====================================================================
1279 1. Put dump sources under CVS, added Id tags in all files so
1280 one can use 'ident' on binary files.
1282 2. Added the dump/restore version in the usage text so one can
1283 easily verify the version he is using.
1285 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
1286 a va_start/va_end problem on linux-ppc (always call va_start
1287 va_end in pairs each time we use a vararg function).
1289 4. Added again the DT_* constants because old libc does not
1290 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
1291 for submitting the bug report.
1293 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
1294 this should enable dumping big (huge) filesystems.
1296 6. Added the RSH environment variable in order to be able to
1297 use a rsh replacement like ssh when doing remote backups (and
1298 bypass the security limitations of rcmd). Now you can do remote
1299 backups without being root (or making dump setuid root).
1301 7. Modified again the way dumpdates works. For incremental dumps,
1302 we need to read dumpdates even if we are not using 'u' option.
1303 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
1306 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
1307 ===================================================================
1309 1. Removed the 'k' flag from the restore 'about' text if kerberos
1310 was not compiled in.
1312 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
1313 to them (fsetflags takes a char*, setflags an open fd!).
1315 3. (f)setflags is called only if the flags aren't empty. If the
1316 file is a special file, a warning is printed, because changing
1317 flags implies opening the device. Normally, a special file
1318 should not have any flag... (Debian bug #29775, patch provided
1319 by Abhijit Dasgupta <abhijit@ans.net>).
1321 4. Made possible to dump a file system not mentioned in /etc/fstab.
1322 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
1324 5. Changed the default behaviour to not create dumpdates
1325 unless 'u' option is specified. Removed the old "debian-patch"
1326 which provided the same thing. (Debian bug #38136, #33818).
1328 6. Removed all those dump*announce, since they were getting old...
1330 7. Added warning messages if dumpdates does not exist and
1331 when an inferior level dump does not exist (except for a level 0
1334 8. Debugged the glob calls in interactive mode: restore used a
1335 dirent struct which was different from the /usr/include/dirent.h
1336 one (this used to work, is it a glibc2 change?), so none of the
1337 compat glob (which used /usr/include/dirent.h) or the system glob
1338 worked. Restore use now the system dirent (and the system
1339 DT_* constants), which are compatible with BSD ones.
1341 9. Added a configure flag (--with-dumpdatespath) to specify
1342 the location of dumpdates file. By default, it is
1345 10. Added the "AUTHOR" and "AVAILABILITY" sections and
1346 included the current date/version in man pages.
1348 11. Corrected the estimation of remaining time when
1349 the operator doesn't change the tapes quickly enough. This
1350 was an old bug, I thought I corrected it, and discovered
1351 that in fact it was corrected in two different places, so
1352 the results canceled each other...
1354 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
1355 ===================================================================
1357 1. Integrated multiple patches from RedHat, Debian and SuSE:
1359 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
1361 - several fixes in the man pages.
1362 - update the default tape device to /dev/st0.
1363 - many updates for Linux Alpha (byte ordering, size_t etc).
1365 - use environment variable for TMPDIR (instead of /tmp).
1366 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
1367 - workaround egcs bug (RedHat bugs #4281 and #2989).
1368 - wire $(OPT) throughout Makefile's.
1370 2. Upgrade the dump revision to 1, making possible to dump filesystems
1371 made with e2fsprogs-1.15 or newer. Nothing seems to break...
1373 3. Fix some compile warnings, prototype all functions.
1375 4. Use glibc err/glob instead of internal compatibility
1376 routines (only if available).
1378 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
1379 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
1381 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
1382 ======================================================================
1384 1. Integrated the changes from FreeBSD-3.1-RELEASE
1385 (mostly bug fixes, buffer overruns, dump has now an "automatic
1386 tape length calculation" flag, dump/restore can use kerberos now
1387 (this is NOT tested), use environment variables for TAPE and
1390 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
1392 3. Corrected a bug in restore when using 'C' option with multi-volumes
1393 tapes (files splited accros two tapes give "size changed" errors
1396 4. Corrected the long standing bug when dumping multiple tapes.
1397 This works for me, needs further testing.
1399 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
1400 ====================================================================
1402 1. Dump now runs correctly on kernels 2.1.x
1403 Fix made by Gerald Peters <gapeters@worldnet.att.net>
1405 Changes between versions 0.4b2 and 0.4b3
1406 ========================================
1408 1. Use realpath() if available
1410 2. Report statistics
1412 Changes between versions 0.4b1 and 0.4b2
1413 ========================================
1415 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
1418 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
1420 3. Fixed configure.in to correctly handle --enable-rmt
1422 Changes between versions 0.3 and 0.4b1
1423 ======================================
1425 1. Integrated the changes from 4.4BSD-Lite2
1427 2. Integrated the patches from Debian and Red Hat
1429 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
1431 4. Changed dump to use the Ext2fs library to get block addresses. This
1432 should solve the endianness problem on SparcLinux.
1434 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
1435 distribution's one) to use autoconf
1437 6. Fixed a few minor bugs
1439 Changes between versions 0.2e and 0.2f
1440 ======================================
1442 1. Added the creation of named pipes (fifos) in restore.
1444 2. Added the -N flag in the restore manual page.
1446 3. Added the file kernel.patch which contains the llseek() optimization
1447 patch for 1.2.x kernels.
1449 4. Fixed a bug in the restoration of symbolic links: owner and group were
1452 5. Integrated some changes from FreeBSD 2.2.
1454 6. Added a call to ftruncate() after restoring each file to restore
1455 correctly files ending by a hole.
1457 Changes between versions 0.2d and 0.2e
1458 ======================================
1460 1. Fixed a bug in the "set owner/mode" process. Every file was restored
1461 with owner = root (0) and group = root/wheel/whatever (0).
1463 Changes between versions 0.2c and 0.2d
1464 ======================================
1466 1. Dump is now able to backup 2GB+ filesystems.
1468 2. Dump and restore can now be linked as static binaries.
1470 Changes between versions 0.2b and 0.2c
1471 ======================================
1473 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
1475 Changes between versions 0.2a and 0.2b
1476 ======================================
1478 1. Really fixed the bug that I should have corrected in 0.2a.
1480 2. Enabled optimization again.
1482 Changes between versions 0.2 and 0.2a
1483 =====================================
1485 1. Disabled the optimization during compilation.
1487 Changes between versions 0.1 and 0.2
1488 ====================================
1490 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
1491 the fs_type field (actually, I modified the file fstab.c to make it
1492 use the mntent functions).
1494 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
1495 made by dump should be readable by the BSD restore and backups made
1496 by the BSD dump should be readable by restore. Unfortunately, this
1497 means that the dump format has changed between version 0.1 and version
1500 3. Dump is now able to backup a subtree, it is no longer limited to whole
1501 filesystems like the BSD version.
1503 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1506 Changes between versions 0.0 and 0.1
1507 ====================================
1509 1. Now create links rdump and rrestore during the `make install' step.
1511 2. Linux port specific bugs added to the manual pages
1513 3. Incorrect estimation of the number of tapes blocks fixed when doing
1514 incremental backups.
1516 4. Better ls-like format in restore in interactive mode.