1 $Id: CHANGES,v 1.147 2001/11/17 10:58:24 stelian Exp $
3 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
4 =======================================================================
6 1. Added a mini howto from Patrick Walsh in the examples
9 2. Minor man pages syntax corrections. Thanks to
10 Chris Danis <screechco@home.com> for reporting the bugs.
12 3. Added a script from David B. Peterson <dave@toppledwagon.com>
13 to the examples section. It features dumping several
14 filesystems to a remote tape drive upon ssh.
16 4. Added a patch provided by Richard Jones <rich@annexia.org>
17 which allows BRADEMAX (number of read errors tolerated by
18 dump) to be adjusted using the -I option.
20 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
21 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
24 6. Add the -l option to restore to specify if, when doing a
25 remote restore, the file used to access the backup is a
26 regular file (the defaults being a tape drive). Restore needs
27 to know this information when reading a remote compressed
28 dump. Previously, this information was autodetected, but
29 the autodetection code fails (with ioctl: Inappropriate ioctl
30 for device) when using a non Linux remote box. Thanks to
31 many users and especially to Eros Albertazzi
32 <eros@lamel.bo.cnr.it> for reporting this.
34 7. Found a workaroung for the dump deadlock problem (3 childs
35 stuck in pause(), father in read()). The workaround seems
36 to work for me and several beta-testers. If it doesn't work
37 for you, please report back.
39 8. Updated the RPM spec file (BuildPrereq, URL etc).
41 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
42 ========================================================================
44 1. Fixed the permissions of a newly created QFA file by dump.
46 2. Cleaned up the source of dump (the external variables
47 definition was a complete mess, making possible to have
50 3. Fixed restore to use the full tape volume path when doing
51 a compare (since it changes the working directory to the
52 filesystem being compared in the process).
54 4. Added the -q option to dump which makes dump abort
55 whenever it needs operator attention. It should be
56 easier to use dump in scripts with this option.
58 5. Detect the use of incompatible options to dump and
59 refuse them (like -a and -B options together).
61 6. Added bzip2 compression to dump/restore (use option -j level
62 to select it). Note that you will need a restore version
63 >= 0.4b24 in order to restore a bzip2 compressed dump.
64 The same warning as for the zlib compression applies:
65 the tape format of a bzip2 dump is not compatible with the
66 original BSD tape format.
68 7. Fixed a overflow problem in dump corrupting the dump when
69 very large files were encountered. Thanks to Vladimir Nadvornik
72 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
73 kernel buffer/page cache before starting the dump, helping
74 a bit those who use dump on mounted filesystems. Thanks to
75 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
78 9. Updated the RPM spec file following the RedHat changes
79 (dynamically linked binaries now in /usr/sbin etc).
81 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
82 which enables restore to recognise multi volume compressed dumps
83 done on CD. Included his scripts for dump (which pipe the dump
84 output directly into cdrecord) and restore. There is now
85 possible to to backups to CD on the fly!
87 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
88 ===================================================================
90 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
91 Marcus Meissner (Caldera International Security Dept.).
93 2. Implement the Sun rmt extended protocol. Patch submitted
94 by Ian Gordon <iangordon@users.sourceforge.net>.
96 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
97 the linux kernel header. This ensures that dump always has the
98 latest version of this file. Patch submitted by
99 Andreas Dilger <adilger@turbolinux.com>.
101 4. Report any filesystem present in either /etc/fstab with a
102 positive passno or /etc/dumpdates in dump -w output.
103 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
105 5. Fixed the looping problem in dump introduced in the
108 6. Changed the -B option of dump to limit the size of
109 _compressed_ output per volume if compression is on.
110 Patch contributed by Helmut Jarausch
111 <jarausch@igpm.rwth-aachen.de>. Note however that, since
112 it is impossible to predict the size of the compressed
113 data before doing the compression, the -B limit is a bit
116 7. Fixed a bug in reading the operator typed file/tape path for
117 the next volume in restore.
119 8. Implemented a "-F script" option to restore which permits the
120 user to specify a script which will be launched at the
121 beginning of each tape, useful for automatic programming of
122 tape changers for example. See the restore man page for the
123 script parameters and return codes.
125 9. Small fix for the QFA routines provided by Uwe Gohlke
126 <uwe@ugsoft.de>, and some recommendations for QFA uses in
129 10. Fixed the multivolume restoring where making a mistake
130 to the 'Mount next tape' prompt caused several blocks to
133 11. Enhanced the -e option of dump to take as a parameter a
134 comma separated list of inode numbers.
136 12. Added the -E option to dump which specify a file containing
137 inode numbers to exclude from the dump.
139 13. Fixed the compressed multi-volume dump + restore.
141 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
142 ==================================================================
144 1. Made dump report the number of blocks dumped per volume.
145 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
147 2. Fix a bug in dump where waiting too much at the 'change volumes'
148 question caused the volume to be rewritten. Thanks to
149 Graham Reed <greed@users.sourceforge.net> for reporting the
150 bug and providing a patch.
152 3. Added a compression option to dump, contributed by Jerry
153 Peters <gapeters@worldnet.att.net>.
155 WARNING: the compression works only on files/pipes and
156 drives supporting variable block size.
158 WARNING2: tapes created using the compression feature are
159 incompatible with the standard dump tape format, and a
160 version of restore >= 0.4b22 will be needed for extraction.
162 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
163 architectures. Thanks to Paul Slootman <paul@debian.org> for
164 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
167 5. Many cleanups (CPP defines, const char warnings, check of
168 ext2fs COMPAT flags, time_t cleanups, added libext2 version
169 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
171 6. Made --prefix option work in configure. All the install path
172 are now based on the configure parameters.
174 7. Added the Quick File Access mode in dump/restore, contributed
175 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
176 a file tape position for each inode, and this file is used by
177 restore (if called with parameter Q and the filename)
178 to directly position the tape at the file restore is currently
179 working on. This saves hours when restoring single files from
180 large backups, saves the tapes and the drive's head. Use
181 --enable-qfa option of configure to compile in the QFA support.
183 8. Added the possibility to dump several files and directories
184 in a single invocation of dump. Thanks to Uwe Gohlke
185 <uwe@ugsoft.de> for implementing this option.
187 9. Fixed the dumping and restoring of files with holes
188 (files ending with holes didn't get dumped or restored
191 10. Fixed a socket descriptor leak in dump, which leaved opened
192 3 file descriptors per dump process (and there is one dump
195 11. Fixed dump large file system support, by explicit use of
196 open64/lseek64/etc functions (explicit use needed because
197 e2fsprogs libraries don't behave well when compiled with
198 FILE_OFFSET_BITS=64).
200 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
201 ======================================================================
203 1. Fixed some bugs in the dump -w|-W logic introduced by
204 the previous version. Thanks to Andreas Dilger
205 <adilger@turbolinux.com> for his help on this one.
207 2. Fixed again a compilation problem when using old e2fs
208 headers (filesystem label related). Thanks to many users
209 who reported this stupid error.
211 3. Fixed a build problem on old lib5 based systems dealing with
212 _PATH_MNTTAB being undefined. Thanks to John Adams
213 <johna@onevista.com> for reporting the bug.
215 4. Improved the error detection in restore when creating
216 the temporary files in TMPDIR. Now restore will corectly
217 report a 'no space left on device' error instead of
218 strange errors which could imply an archive corruption.
219 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
220 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
222 5. Added the throughput information in dump's progression
223 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
226 6. Use libext2fs's inode scanning routines, which are particularly
227 robust when dealing with errors (especially when having some
228 bad blocks on the filesystem). Many thanks to Colin
229 <colin@horizon.com> for reporting the bug, investigating
230 the issues, finding a workaround, writing the patch and
231 fully testing it... (of course, if this patch will break
232 anything, he is to blame for :-)).
234 7. Made dump and restore LFS aware. Dump can dump a filesystem
235 containing large files, generate a large file on output and
236 restore can restore them. This feature is not enabled by
237 default, use --enable-largefile option of configure to enable
238 it (you need to have a LFS aware glibc though). Thanks to
239 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
240 and to Theodore T'so <tytso@valinux.com> for his always
243 8. Made dump ask upon a tape write error if it should rewrite
244 the current volume (assume this is a bad tape) or if it should
245 assume an end-of-tape condition (useful for tape drives which
246 behaves badly at the end of the tape). Thanks to Andreas
247 Dilger <adilger@turbolinux.com> for the suggestion.
249 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
250 =======================================================================
252 1. Fixed a small compilation problem due to a change
253 in the definintion of the struct sigaction in
254 glibc 2.0 / libc5. Thanks to Gunther Schlegel
255 <schlegel@riege.de> for reporting the bug and to
256 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
259 2. Modified the label and uuid analysis in order to be
260 self-contained (independant of kernel/libc headers). This
261 should fix the compile with older kernel/libc headers and
262 will preserve the functionality. Thanks to Bernhard Erdmann
263 <bernhard.erdmann@gmx.de> for reporting the bug.
265 3. The 'exclude inode' option, if applied to a directory
266 inode, excludes now correctly all the leaves of this
267 directory. Thanks to John R. Dennison
268 <gerdesas@users.sourceforge.net> for reporting the bug.
270 4. Fixed the '-e' option to disable the possibility
271 to exclude the root inode (which causes the dumps to
272 be unreadable by restore). Prevented array overflow
273 when multiple -e options are used.
275 5. Fixed dump to correctly interpret a filesystem argument
276 which is a mountpoint and it is not an absolute path
277 (as specified in the fstab). Thanks to Bernhard R. Erdmann
278 <be@berdmann.de> for reporting the bug.
280 6. Made dump able to backup files larger than 2 GB. Note that
281 dump still doesn't cope with files larger than 4 GB.
283 7. Restore the real uid and gid before invoking an external
284 RSH program (big hole when dump or restore is suid root!).
286 8. Get the values for the default tape device and the location
287 of fstab file from the system headers. Thanks to
288 Andreas Dilger <adilger@turbolinux.com> for the patch.
290 9. Made dump -w|-W report all recognized filesystems
291 present in either /etc/fstab or /etc/dumpdates, and present
292 the list in the same order as in fstab file. Thanks
293 to Andreas Dilger <adilger@turbolinux.com> for the patch.
295 10. Made dump's -a (automatic end of tape detection) the
296 default. Specifying one of -B, -c, -d or -s options will
297 override the EOM detection. Thanks to Andreas Dilger
298 <adilger@turbolinux.com> for the patch.
300 11. Save the ext2 filesystem label into the dump volume label.
301 Specifying a volume label on the command line (-L option)
302 will override this feature. Thanks to Andreas Dilger
303 <adilger@turbolinux.com> for the patch.
305 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
306 =====================================================================
308 1. Fixed the signal handling in dump (which I broke in 0.4b17)
309 which was causing several strange dump failures (dump
310 hanged or segmentation faults etc).
312 2. Specified the default blocksize in the dump man page.
314 3. Changed two info messages of restore to be written on stdout
315 instead of stderr in order to leave stderr only for errors
316 or warnings. Thanks to Stephen Atwell
317 <satwell@urbana.css.mot.com> for the suggestion.
319 4. Corrected an off by one calculation which prevented
320 dumping a directory having a 1 character name length.
321 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
322 for reporting the bug.
324 5. Reinforce dump in order to allow only level 0 and no
325 -u option when dumping a subdirectory, and document
326 this in the man page. Thanks to Bernhard Erdmann
327 <bernhard.erdmann@gmx.de> for reporting the bug.
329 6. Fixed a small harmless bug in restore which caused
330 in some conditions hard links to be restored several
331 times (therefore generation some warning messages).
332 Thanks to Kenneth Porter <shiva@well.com> for
335 7. Updated the RPM spec file to the latest RedHat version,
336 providing FHS packaging and other cosmetic changes.
337 You will need RPM version >= 3.0.5 in order to build the RPM.
339 8. Updated the configure script to check for libtermcap
340 before checking for libreadline, because we need this
341 library in order to compile the readline support.
343 9. Made dump understand the LABEL= and UUID= notation
344 both in /etc/fstab and on the command line. Note that
345 you will need the /proc filesystem in order to use
346 these notations. Thanks to Erik Troan <ewt@redhat.com>
347 for providing the patch.
349 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
350 ===================================================================
352 1. Fixed a potential buffer overflow in restore. Thanks
353 to Stan Bubrouski <satan@fastdial.net> for reporting
356 2. Fixed a readline-related bug which prevented
357 'cat DUMPFILE | restore -i -f -' from working. Thanks
358 to Charles Karney <karney@users.sourceforge.net>
361 3. Changed a few "panic" into "exit", causing restore to
362 be more stable against some attacks (like the last one
363 reported on Bugtraq, although the last version of restore
364 was not vulnerable - just dumped core). Thanks to
365 Andreas Hasenack <andreas@conectiva.com.br> for reporting
368 4. Removed the suid-root bit on dump and restore in the
369 default build (and generated RPMs). It should be safer
370 now. Users who need the suid-root capabilities in order
371 to make network backups should read first the man page
372 of dump and enable the suid bit manually.
374 5. Added -ltermcap to the compile parameters for restore
375 when using readline support, in order to make the compile
376 process work on some systems (SuSE ?). Thanks to
377 Patrik Schindler <poc@pocnet.net> for reporting the bug.
379 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
380 ====================================================================
382 1. The -F script is called now *only* at the end of a tape,
383 not at the end of the dump. Thanks to Christian Haul
384 <haul@informatik.tu-darmstadt.de> for the bug report.
386 Normally, the device name and the current volume number
387 are passed on the command line of the script. However,
388 if you want the old style script (with no arguments
389 passed on the command line), you can enable it in
390 configure with the --enable-oldstylefscript.
392 2. Use posix signal handling to preserve dump functionality
393 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
396 3. Made the exit code of restore in case of a 'C'ompare
397 command reflect the result of the compare. An exit status
398 of 0 means the dump archive is correct, 1 means tape errors,
399 2 means that some files were modified. Thanks to Kenneth Porter
400 <shiva@well.com> for the suggestion.
402 4. Made (finally) quotes work in interactive restore.
404 5. Small fixes in order to allow dump to compile with a
405 really old e2fsprogs version. Thanks to Ian Zimmerman
406 <itz@speakeasy.org> for the bug report.
408 6. Add GNU readline capabilities to interactive restore.
409 Use configure's --enable-readline flag to enable this feature.
410 Thanks to Patrik Schindler <poc@pocnet.net> for the
413 7. Do the compare on the fly in restore 'C' mode (this will
414 allow not to exhaust the available /tmp space when
415 ccmparing large files). Thanks to Kenneth Porter
416 <shiva@well.com> for the suggestion.
418 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
419 ====================================================================
421 1. Fixed some several duplicate 'DUMP: DUMP:' in the
424 2. Corrected the estimation of blocks to dump. Note that
425 this is still wrong for large directory inodes, due
426 to the size of a BSD directory entry on the tape
427 (estimation is based on the size of the ext2 directory,
428 which is smaller because it doesn't contain the null
429 character at the end).
431 3. Made dump report the total number of MB written to
432 tape at the end of run. Thanks to W. Reilly Cooley
433 <wcooley@nakedape.cc> for the patch.
435 4. Added the -X option to restore giving the possibility
436 to read the names of the files to be extracted/listed
437 from a text file (in addition of the command line).
438 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
441 5. Added the device name and the current volume number
442 as arguments to the end of tape script (dump -F option).
444 6. Made the multi-volume dump work again (after having
445 broken it in 0.4b15).
447 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
448 ===================================================================
450 1. Added a prompt command in interactive restore mode. Thanks
451 to Andreas Dilger <adilger@home.com> for the patch.
453 2. Fixed a buffer overflow problem in dump (caused by
454 not checking the size of the filesystem parameter).
455 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
456 reporting this on Bugtraq (and to several dump users
457 who forwarded me his mail).
459 3. Added the '-F script' option to dump in order to
460 launch a script at the end of each tape (to be used
461 with a tape changer, or to notify the sysadmin by
464 4. Fixed a bug in restore compare code caused by the changes
467 5. Fixed the treatment of options using the old BSD syntax
468 in both dump and restore.
470 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
471 =======================================================================
473 1. Fixed a bug in dump which may have caused invalid deleted
474 directories to be dumped out if they were deleted after the
475 mapping pass. This could occure on active filesystem and lead
476 to heap corruption (causing dump malfunction in many possible ways).
477 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
478 bug and submitting the patch.
480 2. Fixed the handling of the filesystem parameter in dump. A
481 '/mnt/disk' parameter caused the disk contents to be dumped,
482 but a '/mnt/disk/' parameter caused the mountpoint directory
483 to be dumped (generally an empty directory).
485 3. Improved the output of dump in order to tell which directory
486 it is currently dumping (when dumping a subtree).
488 4. Added the '-e' exclude inode option to dump. Thanks to
489 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
491 5. Added a REPORTING-BUGS file in order to provide a guide
492 on how to correctly report a bug in dump/restore.
494 6. Fixed a restore problem when restoring a hard link to an inode
495 having the immutable or append-only attribute set. Thanks to
496 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
498 7. Fixed a compatibility problem between dumps made on little
499 endian machines (the format was correct) and big endian
500 machines (the format was incorrect). This fix break the
501 compatibility with the older dumps made on big endian
502 machines (sparc, m86k, ppc etc). For the first time in
503 linux dump's history (I believe), the dumps made by *BSD,
504 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
505 so interchangeable. Thanks to Rob Cermak
506 <cermak@ahab.rutgers.edu> for submitting the bug and
507 helping me test the fix.
509 8. Fixed the way dump reports the remaining percent/time, if
510 the number of blocks actually dumped exceeds the estimated
511 number of blocks. Thanks to Jean-Paul van der Jagt
512 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
514 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
515 ======================================================================
517 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
518 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
520 2. Removed the requirement to build the RPM as root from the
521 spec file. Thanks to Christian Weisgerber
522 <naddy@mips.rhein-neckar.de> for submitting this.
524 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
525 causing dump to block when dumping really huge filesystems.
526 Many thanks to Patrik Schindler <poc@pocnet.net> for
527 helping me find this bug.
529 4. Fixed the treatment for an interrupt signal when dump access
530 the remote tape through RSH. Thanks to Christian Weisgerber
531 <naddy@mips.rhein-neckar.de> for providing the patch.
533 5. Fixed a bug which was causing dump/restore to display
534 garbage characters instead of the remote host name.
536 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
537 =====================================================================
539 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
540 <kukuk@suse.de> for submitting the patch.
542 2. Fix for the exit code when using the size estimate option of
543 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
544 submitting the patch.
546 3. Handle EINTR in atomical reads/writes in dump, which was causing
547 dump to fail on some systems. Thanks to Eric Jergensen
548 <eric@dvns.com> for reporting the bug and submitting the patch.
550 4. Handle more than 16 characters for the device names in dumpdates.
551 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
552 tracking down the problem and proposing the solution.
554 5. Fixed a bug in dump which prevented the creation of the
555 dumpdates file when doing a 0-level dump without already
556 having a dumpdates file. Thanks to Patrik Schindler
557 <poc@pocnet.net> for reporting the bug.
559 6. Changed the way dump 'S' flag reports the size estimate
560 from number of blocks into bytes (making it compatible
561 with the Solaris version, and simplifying things for
562 amanda users). Thanks to Jason L Tibbitts III
563 <tibbs@math.uh.edu> for reporting the bug.
565 7. Fixed a compatibility problem in linux/alpha dump tape format.
566 Now the linux/alpha dump are (again) compatible with the
567 other dump formats. But this breaks compatibility with
568 older dumps made on alpha. Thanks to Mike Tibor
569 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
571 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
572 ======================================================================
574 1. Added a '--enable-kerberos' to configure.
576 2. Added a 'S' option to dump which determines the amount of space
577 that is needed to perform the dump without actually doing it, similar
578 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
579 <cermak@ahab.rutgers.edu>.
581 3. Added a 'M' multi-volume option to dump and restore which enables
582 dumping to multiple files (useful when dumping to an ext2
583 partition to make several dump files in order to bypass the 2GB
584 file size limitation). The 'f' argument is treated as a prefix and
585 the output files will be named <prefix>001, <prefix>002 etc. With
586 the 'M' flag, restore automatically selects the right file without
587 asking to enter a new tape each time.
589 4. Fixed a memory leak which was causing dump to grow very big
590 (270MB when dumping a 10GB filesystem...). Thanks to Jason
591 Fearon <jasonf@netrider.org.au> for reporting the bug.
593 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
594 ======================================================================
596 1. Make configure test if the system glob routines support
597 extended syntax (ALTDIRFUNC). If not, use the internal glob
598 routines instead of system ones. Thanks to Bernhard Sadlowski
599 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
600 and helping me resolve this and other minor libc5 compiling
603 2. Fix a problem when dumping a ext2fs with the 'filetype'
604 feature enabled. Thanks to Patrick J. LoPresti
605 <patl@cag.lcs.mit.edu> for reporting the bug and to
606 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
608 3. Made the nodump flag work on directories. A directory which
609 has the nodump flag gets never dumped, regardless of its
612 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
613 which allows dump on an active ext3 filesystem. However, this
614 is a "quick and dirty" patch which enables backup of an ext3
615 filesystem through the ext2 compatibility (by ignoring the
616 NEEDS_RECOVERY bit). The journal file is not recognized and
617 it is dumped (it should not).
619 5. Test the superblock compatibility flags when dumping, in order
620 to be sure that we know how to deal with specific features.
622 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
623 ====================================================================
625 1. Use lchown instead of chown, fixing a possible security problem
626 when restoring symlinks (a malicious user could use this
627 to deliberately corrupt the ownership of important system files).
628 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
629 this and providing the patch.
631 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
632 ====================================================================
634 1. Put dump sources under CVS, added Id tags in all files so
635 one can use 'ident' on binary files.
637 2. Added the dump/restore version in the usage text so one can
638 easily verify the version he is using.
640 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
641 a va_start/va_end problem on linux-ppc (always call va_start
642 va_end in pairs each time we use a vararg function).
644 4. Added again the DT_* constants because old libc does not
645 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
646 for submitting the bug report.
648 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
649 this should enable dumping big (huge) filesystems.
651 6. Added the RSH environment variable in order to be able to
652 use a rsh replacement like ssh when doing remote backups (and
653 bypass the security limitations of rcmd). Now you can do remote
654 backups without being root (or making dump setuid root).
656 7. Modified again the way dumpdates works. For incremental dumps,
657 we need to read dumpdates even if we are not using 'u' option.
658 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
661 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
662 ===================================================================
664 1. Removed the 'k' flag from the restore 'about' text if kerberos
667 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
668 to them (fsetflags takes a char*, setflags an open fd!).
670 3. (f)setflags is called only if the flags aren't empty. If the
671 file is a special file, a warning is printed, because changing
672 flags implies opening the device. Normally, a special file
673 should not have any flag... (Debian bug #29775, patch provided
674 by Abhijit Dasgupta <abhijit@ans.net>).
676 4. Made possible to dump a file system not mentioned in /etc/fstab.
677 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
679 5. Changed the default behaviour to not create dumpdates
680 unless 'u' option is specified. Removed the old "debian-patch"
681 which provided the same thing. (Debian bug #38136, #33818).
683 6. Removed all those dump*announce, since they were getting old...
685 7. Added warning messages if dumpdates does not exist and
686 when an inferior level dump does not exist (except for a level 0
689 8. Debugged the glob calls in interactive mode: restore used a
690 dirent struct which was different from the /usr/include/dirent.h
691 one (this used to work, is it a glibc2 change?), so none of the
692 compat glob (which used /usr/include/dirent.h) or the system glob
693 worked. Restore use now the system dirent (and the system
694 DT_* constants), which are compatible with BSD ones.
696 9. Added a configure flag (--with-dumpdatespath) to specify
697 the location of dumpdates file. By default, it is
700 10. Added the "AUTHOR" and "AVAILABILITY" sections and
701 included the current date/version in man pages.
703 11. Corrected the estimation of remaining time when
704 the operator doesn't change the tapes quickly enough. This
705 was an old bug, I thought I corrected it, and discovered
706 that in fact it was corrected in two different places, so
707 the results canceled each other...
709 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
710 ===================================================================
712 1. Integrated multiple patches from RedHat, Debian and SuSE:
714 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
716 - several fixes in the man pages.
717 - update the default tape device to /dev/st0.
718 - many updates for Linux Alpha (byte ordering, size_t etc).
720 - use environment variable for TMPDIR (instead of /tmp).
721 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
722 - workaround egcs bug (RedHat bugs #4281 and #2989).
723 - wire $(OPT) throughout Makefile's.
725 2. Upgrade the dump revision to 1, making possible to dump filesystems
726 made with e2fsprogs-1.15 or newer. Nothing seems to break...
728 3. Fix some compile warnings, prototype all functions.
730 4. Use glibc err/glob instead of internal compatibility
731 routines (only if available).
733 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
734 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
736 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
737 ======================================================================
739 1. Integrated the changes from FreeBSD-3.1-RELEASE
740 (mostly bug fixes, buffer overruns, dump has now an "automatic
741 tape length calculation" flag, dump/restore can use kerberos now
742 (this is NOT tested), use environment variables for TAPE and
745 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
747 3. Corrected a bug in restore when using 'C' option with multi-volumes
748 tapes (files splited accros two tapes give "size changed" errors
751 4. Corrected the long standing bug when dumping multiple tapes.
752 This works for me, needs further testing.
754 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
755 ====================================================================
757 1. Dump now runs correctly on kernels 2.1.x
758 Fix made by Gerald Peters <gapeters@worldnet.att.net>
760 Changes between versions 0.4b2 and 0.4b3
761 ========================================
763 1. Use realpath() if available
767 Changes between versions 0.4b1 and 0.4b2
768 ========================================
770 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
773 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
775 3. Fixed configure.in to correctly handle --enable-rmt
777 Changes between versions 0.3 and 0.4b1
778 ======================================
780 1. Integrated the changes from 4.4BSD-Lite2
782 2. Integrated the patches from Debian and Red Hat
784 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
786 4. Changed dump to use the Ext2fs library to get block addresses. This
787 should solve the endianness problem on SparcLinux.
789 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
790 distribution's one) to use autoconf
792 6. Fixed a few minor bugs
794 Changes between versions 0.2e and 0.2f
795 ======================================
797 1. Added the creation of named pipes (fifos) in restore.
799 2. Added the -N flag in the restore manual page.
801 3. Added the file kernel.patch which contains the llseek() optimization
802 patch for 1.2.x kernels.
804 4. Fixed a bug in the restoration of symbolic links: owner and group were
807 5. Integrated some changes from FreeBSD 2.2.
809 6. Added a call to ftruncate() after restoring each file to restore
810 correctly files ending by a hole.
812 Changes between versions 0.2d and 0.2e
813 ======================================
815 1. Fixed a bug in the "set owner/mode" process. Every file was restored
816 with owner = root (0) and group = root/wheel/whatever (0).
818 Changes between versions 0.2c and 0.2d
819 ======================================
821 1. Dump is now able to backup 2GB+ filesystems.
823 2. Dump and restore can now be linked as static binaries.
825 Changes between versions 0.2b and 0.2c
826 ======================================
828 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
830 Changes between versions 0.2a and 0.2b
831 ======================================
833 1. Really fixed the bug that I should have corrected in 0.2a.
835 2. Enabled optimization again.
837 Changes between versions 0.2 and 0.2a
838 =====================================
840 1. Disabled the optimization during compilation.
842 Changes between versions 0.1 and 0.2
843 ====================================
845 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
846 the fs_type field (actually, I modified the file fstab.c to make it
847 use the mntent functions).
849 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
850 made by dump should be readable by the BSD restore and backups made
851 by the BSD dump should be readable by restore. Unfortunately, this
852 means that the dump format has changed between version 0.1 and version
855 3. Dump is now able to backup a subtree, it is no longer limited to whole
856 filesystems like the BSD version.
858 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
861 Changes between versions 0.0 and 0.1
862 ====================================
864 1. Now create links rdump and rrestore during the `make install' step.
866 2. Linux port specific bugs added to the manual pages
868 3. Incorrect estimation of the number of tapes blocks fixed when doing
871 4. Better ls-like format in restore in interactive mode.