1 $Id: CHANGES,v 1.130 2001/08/13 16:17:52 stelian Exp $
3 Changes between versions 0.4b23 and 0.4b24 (released ?????????????)
4 ===================================================================
6 1. Fixed the permissions of a newly created QFA file by dump.
8 2. Cleaned up the source of dump (the external variables
9 definition was a complete mess, making possible to have
12 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
13 ===================================================================
15 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
16 Marcus Meissner (Caldera International Security Dept.).
18 2. Implement the Sun rmt extended protocol. Patch submitted
19 by Ian Gordon <iangordon@users.sourceforge.net>.
21 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
22 the linux kernel header. This ensures that dump always has the
23 latest version of this file. Patch submitted by
24 Andreas Dilger <adilger@turbolinux.com>.
26 4. Report any filesystem present in either /etc/fstab with a
27 positive passno or /etc/dumpdates in dump -w output.
28 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
30 5. Fixed the looping problem in dump introduced in the
33 6. Changed the -B option of dump to limit the size of
34 _compressed_ output per volume if compression is on.
35 Patch contributed by Helmut Jarausch
36 <jarausch@igpm.rwth-aachen.de>. Note however that, since
37 it is impossible to predict the size of the compressed
38 data before doing the compression, the -B limit is a bit
41 7. Fixed a bug in reading the operator typed file/tape path for
42 the next volume in restore.
44 8. Implemented a "-F script" option to restore which permits the
45 user to specify a script which will be launched at the
46 beginning of each tape, useful for automatic programming of
47 tape changers for example. See the restore man page for the
48 script parameters and return codes.
50 9. Small fix for the QFA routines provided by Uwe Gohlke
51 <uwe@ugsoft.de>, and some recommendations for QFA uses in
54 10. Fixed the multivolume restoring where making a mistake
55 to the 'Mount next tape' prompt caused several blocks to
58 11. Enhanced the -e option of dump to take as a parameter a
59 comma separated list of inode numbers.
61 12. Added the -E option to dump which specify a file containing
62 inode numbers to exclude from the dump.
64 13. Fixed the compressed multi-volume dump + restore.
66 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
67 ==================================================================
69 1. Made dump report the number of blocks dumped per volume.
70 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
72 2. Fix a bug in dump where waiting too much at the 'change volumes'
73 question caused the volume to be rewritten. Thanks to
74 Graham Reed <greed@users.sourceforge.net> for reporting the
75 bug and providing a patch.
77 3. Added a compression option to dump, contributed by Jerry
78 Peters <gapeters@worldnet.att.net>.
80 WARNING: the compression works only on files/pipes and
81 drives supporting variable block size.
83 WARNING2: tapes created using the compression feature are
84 incompatible with the standard dump tape format, and a
85 version of restore >= 0.4b22 will be needed for extraction.
87 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
88 architectures. Thanks to Paul Slootman <paul@debian.org> for
89 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
92 5. Many cleanups (CPP defines, const char warnings, check of
93 ext2fs COMPAT flags, time_t cleanups, added libext2 version
94 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
96 6. Made --prefix option work in configure. All the install path
97 are now based on the configure parameters.
99 7. Added the Quick File Access mode in dump/restore, contributed
100 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
101 a file tape position for each inode, and this file is used by
102 restore (if called with parameter Q and the filename)
103 to directly position the tape at the file restore is currently
104 working on. This saves hours when restoring single files from
105 large backups, saves the tapes and the drive's head. Use
106 --enable-qfa option of configure to compile in the QFA support.
108 8. Added the possibility to dump several files and directories
109 in a single invocation of dump. Thanks to Uwe Gohlke
110 <uwe@ugsoft.de> for implementing this option.
112 9. Fixed the dumping and restoring of files with holes
113 (files ending with holes didn't get dumped or restored
116 10. Fixed a socket descriptor leak in dump, which leaved opened
117 3 file descriptors per dump process (and there is one dump
120 11. Fixed dump large file system support, by explicit use of
121 open64/lseek64/etc functions (explicit use needed because
122 e2fsprogs libraries don't behave well when compiled with
123 FILE_OFFSET_BITS=64).
125 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
126 ======================================================================
128 1. Fixed some bugs in the dump -w|-W logic introduced by
129 the previous version. Thanks to Andreas Dilger
130 <adilger@turbolinux.com> for his help on this one.
132 2. Fixed again a compilation problem when using old e2fs
133 headers (filesystem label related). Thanks to many users
134 who reported this stupid error.
136 3. Fixed a build problem on old lib5 based systems dealing with
137 _PATH_MNTTAB being undefined. Thanks to John Adams
138 <johna@onevista.com> for reporting the bug.
140 4. Improved the error detection in restore when creating
141 the temporary files in TMPDIR. Now restore will corectly
142 report a 'no space left on device' error instead of
143 strange errors which could imply an archive corruption.
144 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
145 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
147 5. Added the throughput information in dump's progression
148 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
151 6. Use libext2fs's inode scanning routines, which are particularly
152 robust when dealing with errors (especially when having some
153 bad blocks on the filesystem). Many thanks to Colin
154 <colin@horizon.com> for reporting the bug, investigating
155 the issues, finding a workaround, writing the patch and
156 fully testing it... (of course, if this patch will break
157 anything, he is to blame for :-)).
159 7. Made dump and restore LFS aware. Dump can dump a filesystem
160 containing large files, generate a large file on output and
161 restore can restore them. This feature is not enabled by
162 default, use --enable-largefile option of configure to enable
163 it (you need to have a LFS aware glibc though). Thanks to
164 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
165 and to Theodore T'so <tytso@valinux.com> for his always
168 8. Made dump ask upon a tape write error if it should rewrite
169 the current volume (assume this is a bad tape) or if it should
170 assume an end-of-tape condition (useful for tape drives which
171 behaves badly at the end of the tape). Thanks to Andreas
172 Dilger <adilger@turbolinux.com> for the suggestion.
174 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
175 =======================================================================
177 1. Fixed a small compilation problem due to a change
178 in the definintion of the struct sigaction in
179 glibc 2.0 / libc5. Thanks to Gunther Schlegel
180 <schlegel@riege.de> for reporting the bug and to
181 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
184 2. Modified the label and uuid analysis in order to be
185 self-contained (independant of kernel/libc headers). This
186 should fix the compile with older kernel/libc headers and
187 will preserve the functionality. Thanks to Bernhard Erdmann
188 <bernhard.erdmann@gmx.de> for reporting the bug.
190 3. The 'exclude inode' option, if applied to a directory
191 inode, excludes now correctly all the leaves of this
192 directory. Thanks to John R. Dennison
193 <gerdesas@users.sourceforge.net> for reporting the bug.
195 4. Fixed the '-e' option to disable the possibility
196 to exclude the root inode (which causes the dumps to
197 be unreadable by restore). Prevented array overflow
198 when multiple -e options are used.
200 5. Fixed dump to correctly interpret a filesystem argument
201 which is a mountpoint and it is not an absolute path
202 (as specified in the fstab). Thanks to Bernhard R. Erdmann
203 <be@berdmann.de> for reporting the bug.
205 6. Made dump able to backup files larger than 2 GB. Note that
206 dump still doesn't cope with files larger than 4 GB.
208 7. Restore the real uid and gid before invoking an external
209 RSH program (big hole when dump or restore is suid root!).
211 8. Get the values for the default tape device and the location
212 of fstab file from the system headers. Thanks to
213 Andreas Dilger <adilger@turbolinux.com> for the patch.
215 9. Made dump -w|-W report all recognized filesystems
216 present in either /etc/fstab or /etc/dumpdates, and present
217 the list in the same order as in fstab file. Thanks
218 to Andreas Dilger <adilger@turbolinux.com> for the patch.
220 10. Made dump's -a (automatic end of tape detection) the
221 default. Specifying one of -B, -c, -d or -s options will
222 override the EOM detection. Thanks to Andreas Dilger
223 <adilger@turbolinux.com> for the patch.
225 11. Save the ext2 filesystem label into the dump volume label.
226 Specifying a volume label on the command line (-L option)
227 will override this feature. Thanks to Andreas Dilger
228 <adilger@turbolinux.com> for the patch.
230 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
231 =====================================================================
233 1. Fixed the signal handling in dump (which I broke in 0.4b17)
234 which was causing several strange dump failures (dump
235 hanged or segmentation faults etc).
237 2. Specified the default blocksize in the dump man page.
239 3. Changed two info messages of restore to be written on stdout
240 instead of stderr in order to leave stderr only for errors
241 or warnings. Thanks to Stephen Atwell
242 <satwell@urbana.css.mot.com> for the suggestion.
244 4. Corrected an off by one calculation which prevented
245 dumping a directory having a 1 character name length.
246 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
247 for reporting the bug.
249 5. Reinforce dump in order to allow only level 0 and no
250 -u option when dumping a subdirectory, and document
251 this in the man page. Thanks to Bernhard Erdmann
252 <bernhard.erdmann@gmx.de> for reporting the bug.
254 6. Fixed a small harmless bug in restore which caused
255 in some conditions hard links to be restored several
256 times (therefore generation some warning messages).
257 Thanks to Kenneth Porter <shiva@well.com> for
260 7. Updated the RPM spec file to the latest RedHat version,
261 providing FHS packaging and other cosmetic changes.
262 You will need RPM version >= 3.0.5 in order to build the RPM.
264 8. Updated the configure script to check for libtermcap
265 before checking for libreadline, because we need this
266 library in order to compile the readline support.
268 9. Made dump understand the LABEL= and UUID= notation
269 both in /etc/fstab and on the command line. Note that
270 you will need the /proc filesystem in order to use
271 these notations. Thanks to Erik Troan <ewt@redhat.com>
272 for providing the patch.
274 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
275 ===================================================================
277 1. Fixed a potential buffer overflow in restore. Thanks
278 to Stan Bubrouski <satan@fastdial.net> for reporting
281 2. Fixed a readline-related bug which prevented
282 'cat DUMPFILE | restore -i -f -' from working. Thanks
283 to Charles Karney <karney@users.sourceforge.net>
286 3. Changed a few "panic" into "exit", causing restore to
287 be more stable against some attacks (like the last one
288 reported on Bugtraq, although the last version of restore
289 was not vulnerable - just dumped core). Thanks to
290 Andreas Hasenack <andreas@conectiva.com.br> for reporting
293 4. Removed the suid-root bit on dump and restore in the
294 default build (and generated RPMs). It should be safer
295 now. Users who need the suid-root capabilities in order
296 to make network backups should read first the man page
297 of dump and enable the suid bit manually.
299 5. Added -ltermcap to the compile parameters for restore
300 when using readline support, in order to make the compile
301 process work on some systems (SuSE ?). Thanks to
302 Patrik Schindler <poc@pocnet.net> for reporting the bug.
304 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
305 ====================================================================
307 1. The -F script is called now *only* at the end of a tape,
308 not at the end of the dump. Thanks to Christian Haul
309 <haul@informatik.tu-darmstadt.de> for the bug report.
311 Normally, the device name and the current volume number
312 are passed on the command line of the script. However,
313 if you want the old style script (with no arguments
314 passed on the command line), you can enable it in
315 configure with the --enable-oldstylefscript.
317 2. Use posix signal handling to preserve dump functionality
318 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
321 3. Made the exit code of restore in case of a 'C'ompare
322 command reflect the result of the compare. An exit status
323 of 0 means the dump archive is correct, 1 means tape errors,
324 2 means that some files were modified. Thanks to Kenneth Porter
325 <shiva@well.com> for the suggestion.
327 4. Made (finally) quotes work in interactive restore.
329 5. Small fixes in order to allow dump to compile with a
330 really old e2fsprogs version. Thanks to Ian Zimmerman
331 <itz@speakeasy.org> for the bug report.
333 6. Add GNU readline capabilities to interactive restore.
334 Use configure's --enable-readline flag to enable this feature.
335 Thanks to Patrik Schindler <poc@pocnet.net> for the
338 7. Do the compare on the fly in restore 'C' mode (this will
339 allow not to exhaust the available /tmp space when
340 ccmparing large files). Thanks to Kenneth Porter
341 <shiva@well.com> for the suggestion.
343 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
344 ====================================================================
346 1. Fixed some several duplicate 'DUMP: DUMP:' in the
349 2. Corrected the estimation of blocks to dump. Note that
350 this is still wrong for large directory inodes, due
351 to the size of a BSD directory entry on the tape
352 (estimation is based on the size of the ext2 directory,
353 which is smaller because it doesn't contain the null
354 character at the end).
356 3. Made dump report the total number of MB written to
357 tape at the end of run. Thanks to W. Reilly Cooley
358 <wcooley@nakedape.cc> for the patch.
360 4. Added the -X option to restore giving the possibility
361 to read the names of the files to be extracted/listed
362 from a text file (in addition of the command line).
363 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
366 5. Added the device name and the current volume number
367 as arguments to the end of tape script (dump -F option).
369 6. Made the multi-volume dump work again (after having
370 broken it in 0.4b15).
372 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
373 ===================================================================
375 1. Added a prompt command in interactive restore mode. Thanks
376 to Andreas Dilger <adilger@home.com> for the patch.
378 2. Fixed a buffer overflow problem in dump (caused by
379 not checking the size of the filesystem parameter).
380 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
381 reporting this on Bugtraq (and to several dump users
382 who forwarded me his mail).
384 3. Added the '-F script' option to dump in order to
385 launch a script at the end of each tape (to be used
386 with a tape changer, or to notify the sysadmin by
389 4. Fixed a bug in restore compare code caused by the changes
392 5. Fixed the treatment of options using the old BSD syntax
393 in both dump and restore.
395 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
396 =======================================================================
398 1. Fixed a bug in dump which may have caused invalid deleted
399 directories to be dumped out if they were deleted after the
400 mapping pass. This could occure on active filesystem and lead
401 to heap corruption (causing dump malfunction in many possible ways).
402 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
403 bug and submitting the patch.
405 2. Fixed the handling of the filesystem parameter in dump. A
406 '/mnt/disk' parameter caused the disk contents to be dumped,
407 but a '/mnt/disk/' parameter caused the mountpoint directory
408 to be dumped (generally an empty directory).
410 3. Improved the output of dump in order to tell which directory
411 it is currently dumping (when dumping a subtree).
413 4. Added the '-e' exclude inode option to dump. Thanks to
414 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
416 5. Added a REPORTING-BUGS file in order to provide a guide
417 on how to correctly report a bug in dump/restore.
419 6. Fixed a restore problem when restoring a hard link to an inode
420 having the immutable or append-only attribute set. Thanks to
421 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
423 7. Fixed a compatibility problem between dumps made on little
424 endian machines (the format was correct) and big endian
425 machines (the format was incorrect). This fix break the
426 compatibility with the older dumps made on big endian
427 machines (sparc, m86k, ppc etc). For the first time in
428 linux dump's history (I believe), the dumps made by *BSD,
429 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
430 so interchangeable. Thanks to Rob Cermak
431 <cermak@ahab.rutgers.edu> for submitting the bug and
432 helping me test the fix.
434 8. Fixed the way dump reports the remaining percent/time, if
435 the number of blocks actually dumped exceeds the estimated
436 number of blocks. Thanks to Jean-Paul van der Jagt
437 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
439 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
440 ======================================================================
442 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
443 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
445 2. Removed the requirement to build the RPM as root from the
446 spec file. Thanks to Christian Weisgerber
447 <naddy@mips.rhein-neckar.de> for submitting this.
449 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
450 causing dump to block when dumping really huge filesystems.
451 Many thanks to Patrik Schindler <poc@pocnet.net> for
452 helping me find this bug.
454 4. Fixed the treatment for an interrupt signal when dump access
455 the remote tape through RSH. Thanks to Christian Weisgerber
456 <naddy@mips.rhein-neckar.de> for providing the patch.
458 5. Fixed a bug which was causing dump/restore to display
459 garbage characters instead of the remote host name.
461 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
462 =====================================================================
464 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
465 <kukuk@suse.de> for submitting the patch.
467 2. Fix for the exit code when using the size estimate option of
468 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
469 submitting the patch.
471 3. Handle EINTR in atomical reads/writes in dump, which was causing
472 dump to fail on some systems. Thanks to Eric Jergensen
473 <eric@dvns.com> for reporting the bug and submitting the patch.
475 4. Handle more than 16 characters for the device names in dumpdates.
476 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
477 tracking down the problem and proposing the solution.
479 5. Fixed a bug in dump which prevented the creation of the
480 dumpdates file when doing a 0-level dump without already
481 having a dumpdates file. Thanks to Patrik Schindler
482 <poc@pocnet.net> for reporting the bug.
484 6. Changed the way dump 'S' flag reports the size estimate
485 from number of blocks into bytes (making it compatible
486 with the Solaris version, and simplifying things for
487 amanda users). Thanks to Jason L Tibbitts III
488 <tibbs@math.uh.edu> for reporting the bug.
490 7. Fixed a compatibility problem in linux/alpha dump tape format.
491 Now the linux/alpha dump are (again) compatible with the
492 other dump formats. But this breaks compatibility with
493 older dumps made on alpha. Thanks to Mike Tibor
494 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
496 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
497 ======================================================================
499 1. Added a '--enable-kerberos' to configure.
501 2. Added a 'S' option to dump which determines the amount of space
502 that is needed to perform the dump without actually doing it, similar
503 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
504 <cermak@ahab.rutgers.edu>.
506 3. Added a 'M' multi-volume option to dump and restore which enables
507 dumping to multiple files (useful when dumping to an ext2
508 partition to make several dump files in order to bypass the 2GB
509 file size limitation). The 'f' argument is treated as a prefix and
510 the output files will be named <prefix>001, <prefix>002 etc. With
511 the 'M' flag, restore automatically selects the right file without
512 asking to enter a new tape each time.
514 4. Fixed a memory leak which was causing dump to grow very big
515 (270MB when dumping a 10GB filesystem...). Thanks to Jason
516 Fearon <jasonf@netrider.org.au> for reporting the bug.
518 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
519 ======================================================================
521 1. Make configure test if the system glob routines support
522 extended syntax (ALTDIRFUNC). If not, use the internal glob
523 routines instead of system ones. Thanks to Bernhard Sadlowski
524 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
525 and helping me resolve this and other minor libc5 compiling
528 2. Fix a problem when dumping a ext2fs with the 'filetype'
529 feature enabled. Thanks to Patrick J. LoPresti
530 <patl@cag.lcs.mit.edu> for reporting the bug and to
531 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
533 3. Made the nodump flag work on directories. A directory which
534 has the nodump flag gets never dumped, regardless of its
537 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
538 which allows dump on an active ext3 filesystem. However, this
539 is a "quick and dirty" patch which enables backup of an ext3
540 filesystem through the ext2 compatibility (by ignoring the
541 NEEDS_RECOVERY bit). The journal file is not recognized and
542 it is dumped (it should not).
544 5. Test the superblock compatibility flags when dumping, in order
545 to be sure that we know how to deal with specific features.
547 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
548 ====================================================================
550 1. Use lchown instead of chown, fixing a possible security problem
551 when restoring symlinks (a malicious user could use this
552 to deliberately corrupt the ownership of important system files).
553 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
554 this and providing the patch.
556 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
557 ====================================================================
559 1. Put dump sources under CVS, added Id tags in all files so
560 one can use 'ident' on binary files.
562 2. Added the dump/restore version in the usage text so one can
563 easily verify the version he is using.
565 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
566 a va_start/va_end problem on linux-ppc (always call va_start
567 va_end in pairs each time we use a vararg function).
569 4. Added again the DT_* constants because old libc does not
570 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
571 for submitting the bug report.
573 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
574 this should enable dumping big (huge) filesystems.
576 6. Added the RSH environment variable in order to be able to
577 use a rsh replacement like ssh when doing remote backups (and
578 bypass the security limitations of rcmd). Now you can do remote
579 backups without being root (or making dump setuid root).
581 7. Modified again the way dumpdates works. For incremental dumps,
582 we need to read dumpdates even if we are not using 'u' option.
583 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
586 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
587 ===================================================================
589 1. Removed the 'k' flag from the restore 'about' text if kerberos
592 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
593 to them (fsetflags takes a char*, setflags an open fd!).
595 3. (f)setflags is called only if the flags aren't empty. If the
596 file is a special file, a warning is printed, because changing
597 flags implies opening the device. Normally, a special file
598 should not have any flag... (Debian bug #29775, patch provided
599 by Abhijit Dasgupta <abhijit@ans.net>).
601 4. Made possible to dump a file system not mentioned in /etc/fstab.
602 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
604 5. Changed the default behaviour to not create dumpdates
605 unless 'u' option is specified. Removed the old "debian-patch"
606 which provided the same thing. (Debian bug #38136, #33818).
608 6. Removed all those dump*announce, since they were getting old...
610 7. Added warning messages if dumpdates does not exist and
611 when an inferior level dump does not exist (except for a level 0
614 8. Debugged the glob calls in interactive mode: restore used a
615 dirent struct which was different from the /usr/include/dirent.h
616 one (this used to work, is it a glibc2 change?), so none of the
617 compat glob (which used /usr/include/dirent.h) or the system glob
618 worked. Restore use now the system dirent (and the system
619 DT_* constants), which are compatible with BSD ones.
621 9. Added a configure flag (--with-dumpdatespath) to specify
622 the location of dumpdates file. By default, it is
625 10. Added the "AUTHOR" and "AVAILABILITY" sections and
626 included the current date/version in man pages.
628 11. Corrected the estimation of remaining time when
629 the operator doesn't change the tapes quickly enough. This
630 was an old bug, I thought I corrected it, and discovered
631 that in fact it was corrected in two different places, so
632 the results canceled each other...
634 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
635 ===================================================================
637 1. Integrated multiple patches from RedHat, Debian and SuSE:
639 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
641 - several fixes in the man pages.
642 - update the default tape device to /dev/st0.
643 - many updates for Linux Alpha (byte ordering, size_t etc).
645 - use environment variable for TMPDIR (instead of /tmp).
646 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
647 - workaround egcs bug (RedHat bugs #4281 and #2989).
648 - wire $(OPT) throughout Makefile's.
650 2. Upgrade the dump revision to 1, making possible to dump filesystems
651 made with e2fsprogs-1.15 or newer. Nothing seems to break...
653 3. Fix some compile warnings, prototype all functions.
655 4. Use glibc err/glob instead of internal compatibility
656 routines (only if available).
658 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
659 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
661 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
662 ======================================================================
664 1. Integrated the changes from FreeBSD-3.1-RELEASE
665 (mostly bug fixes, buffer overruns, dump has now an "automatic
666 tape length calculation" flag, dump/restore can use kerberos now
667 (this is NOT tested), use environment variables for TAPE and
670 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
672 3. Corrected a bug in restore when using 'C' option with multi-volumes
673 tapes (files splited accros two tapes give "size changed" errors
676 4. Corrected the long standing bug when dumping multiple tapes.
677 This works for me, needs further testing.
679 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
680 ====================================================================
682 1. Dump now runs correctly on kernels 2.1.x
683 Fix made by Gerald Peters <gapeters@worldnet.att.net>
685 Changes between versions 0.4b2 and 0.4b3
686 ========================================
688 1. Use realpath() if available
692 Changes between versions 0.4b1 and 0.4b2
693 ========================================
695 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
698 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
700 3. Fixed configure.in to correctly handle --enable-rmt
702 Changes between versions 0.3 and 0.4b1
703 ======================================
705 1. Integrated the changes from 4.4BSD-Lite2
707 2. Integrated the patches from Debian and Red Hat
709 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
711 4. Changed dump to use the Ext2fs library to get block addresses. This
712 should solve the endianness problem on SparcLinux.
714 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
715 distribution's one) to use autoconf
717 6. Fixed a few minor bugs
719 Changes between versions 0.2e and 0.2f
720 ======================================
722 1. Added the creation of named pipes (fifos) in restore.
724 2. Added the -N flag in the restore manual page.
726 3. Added the file kernel.patch which contains the llseek() optimization
727 patch for 1.2.x kernels.
729 4. Fixed a bug in the restoration of symbolic links: owner and group were
732 5. Integrated some changes from FreeBSD 2.2.
734 6. Added a call to ftruncate() after restoring each file to restore
735 correctly files ending by a hole.
737 Changes between versions 0.2d and 0.2e
738 ======================================
740 1. Fixed a bug in the "set owner/mode" process. Every file was restored
741 with owner = root (0) and group = root/wheel/whatever (0).
743 Changes between versions 0.2c and 0.2d
744 ======================================
746 1. Dump is now able to backup 2GB+ filesystems.
748 2. Dump and restore can now be linked as static binaries.
750 Changes between versions 0.2b and 0.2c
751 ======================================
753 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
755 Changes between versions 0.2a and 0.2b
756 ======================================
758 1. Really fixed the bug that I should have corrected in 0.2a.
760 2. Enabled optimization again.
762 Changes between versions 0.2 and 0.2a
763 =====================================
765 1. Disabled the optimization during compilation.
767 Changes between versions 0.1 and 0.2
768 ====================================
770 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
771 the fs_type field (actually, I modified the file fstab.c to make it
772 use the mntent functions).
774 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
775 made by dump should be readable by the BSD restore and backups made
776 by the BSD dump should be readable by restore. Unfortunately, this
777 means that the dump format has changed between version 0.1 and version
780 3. Dump is now able to backup a subtree, it is no longer limited to whole
781 filesystems like the BSD version.
783 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
786 Changes between versions 0.0 and 0.1
787 ====================================
789 1. Now create links rdump and rrestore during the `make install' step.
791 2. Linux port specific bugs added to the manual pages
793 3. Incorrect estimation of the number of tapes blocks fixed when doing
796 4. Better ls-like format in restore in interactive mode.