1 $Id: CHANGES,v 1.155 2002/01/11 08:54:12 stelian Exp $
3 Changes between versions 0.4b26 and 0.4b27 (released ???????????????)
4 =====================================================================
6 1. Fixed behaviour of dump when exceeding resource limits
9 2. Added the -L flag to restore to allow the user to specify a
10 maximal allowed number of miscompares when using restore
11 with the -C option to check the backup.
13 3. Detailed the manual entry for the -N option of restore.
15 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
16 =====================================================================
18 1. Added a set of backup scripts from Eugenio Diaz
19 <getnito@yahoo.com> in the examples section. It features
20 automatic (cron based) full and incremental dumping of
21 several filesystems on a separate filesystem.
23 2. Fixed a off-by-one miscalculation which disabled dumping
24 a one letter subdirectory. Thanks to Chris Siebenmann
25 <cks@utcc.utoronto.ca> for reporting the bug.
27 3. Fixed several restore bugs occuring when trying to
28 restore some missing files on the tape. Thanks to Chris
29 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
31 4. Fixed --with-ldopts configure argument passing, installing from
32 a separate object directory, makefile cleanups contributed
33 by <splite@purdue.edu>.
35 5. Fix a bug which could caused, in some conditions, the highest
36 number inode of a filesystem, to not be dumped. Many thanks
37 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
40 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
41 =======================================================================
43 1. Added a mini howto from Patrick Walsh in the examples
46 2. Minor man pages syntax corrections. Thanks to
47 Chris Danis <screechco@home.com> for reporting the bugs.
49 3. Added a script from David B. Peterson <dave@toppledwagon.com>
50 to the examples section. It features dumping several
51 filesystems to a remote tape drive upon ssh.
53 4. Added a patch provided by Richard Jones <rich@annexia.org>
54 which allows BRADEMAX (number of read errors tolerated by
55 dump) to be adjusted using the -I option.
57 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
58 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
61 6. Add the -l option to restore to specify if, when doing a
62 remote restore, the file used to access the backup is a
63 regular file (the defaults being a tape drive). Restore needs
64 to know this information when reading a remote compressed
65 dump. Previously, this information was autodetected, but
66 the autodetection code fails (with ioctl: Inappropriate ioctl
67 for device) when using a non Linux remote box. Thanks to
68 many users and especially to Eros Albertazzi
69 <eros@lamel.bo.cnr.it> for reporting this.
71 7. Found a workaroung for the dump deadlock problem (3 childs
72 stuck in pause(), father in read()). The workaround seems
73 to work for me and several beta-testers. If it doesn't work
74 for you, please report back.
76 8. Updated the RPM spec file (BuildPrereq, URL etc).
78 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
79 ========================================================================
81 1. Fixed the permissions of a newly created QFA file by dump.
83 2. Cleaned up the source of dump (the external variables
84 definition was a complete mess, making possible to have
87 3. Fixed restore to use the full tape volume path when doing
88 a compare (since it changes the working directory to the
89 filesystem being compared in the process).
91 4. Added the -q option to dump which makes dump abort
92 whenever it needs operator attention. It should be
93 easier to use dump in scripts with this option.
95 5. Detect the use of incompatible options to dump and
96 refuse them (like -a and -B options together).
98 6. Added bzip2 compression to dump/restore (use option -j level
99 to select it). Note that you will need a restore version
100 >= 0.4b24 in order to restore a bzip2 compressed dump.
101 The same warning as for the zlib compression applies:
102 the tape format of a bzip2 dump is not compatible with the
103 original BSD tape format.
105 7. Fixed a overflow problem in dump corrupting the dump when
106 very large files were encountered. Thanks to Vladimir Nadvornik
109 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
110 kernel buffer/page cache before starting the dump, helping
111 a bit those who use dump on mounted filesystems. Thanks to
112 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
115 9. Updated the RPM spec file following the RedHat changes
116 (dynamically linked binaries now in /usr/sbin etc).
118 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
119 which enables restore to recognise multi volume compressed dumps
120 done on CD. Included his scripts for dump (which pipe the dump
121 output directly into cdrecord) and restore. There is now
122 possible to to backups to CD on the fly!
124 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
125 ===================================================================
127 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
128 Marcus Meissner (Caldera International Security Dept.).
130 2. Implement the Sun rmt extended protocol. Patch submitted
131 by Ian Gordon <iangordon@users.sourceforge.net>.
133 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
134 the linux kernel header. This ensures that dump always has the
135 latest version of this file. Patch submitted by
136 Andreas Dilger <adilger@turbolinux.com>.
138 4. Report any filesystem present in either /etc/fstab with a
139 positive passno or /etc/dumpdates in dump -w output.
140 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
142 5. Fixed the looping problem in dump introduced in the
145 6. Changed the -B option of dump to limit the size of
146 _compressed_ output per volume if compression is on.
147 Patch contributed by Helmut Jarausch
148 <jarausch@igpm.rwth-aachen.de>. Note however that, since
149 it is impossible to predict the size of the compressed
150 data before doing the compression, the -B limit is a bit
153 7. Fixed a bug in reading the operator typed file/tape path for
154 the next volume in restore.
156 8. Implemented a "-F script" option to restore which permits the
157 user to specify a script which will be launched at the
158 beginning of each tape, useful for automatic programming of
159 tape changers for example. See the restore man page for the
160 script parameters and return codes.
162 9. Small fix for the QFA routines provided by Uwe Gohlke
163 <uwe@ugsoft.de>, and some recommendations for QFA uses in
166 10. Fixed the multivolume restoring where making a mistake
167 to the 'Mount next tape' prompt caused several blocks to
170 11. Enhanced the -e option of dump to take as a parameter a
171 comma separated list of inode numbers.
173 12. Added the -E option to dump which specify a file containing
174 inode numbers to exclude from the dump.
176 13. Fixed the compressed multi-volume dump + restore.
178 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
179 ==================================================================
181 1. Made dump report the number of blocks dumped per volume.
182 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
184 2. Fix a bug in dump where waiting too much at the 'change volumes'
185 question caused the volume to be rewritten. Thanks to
186 Graham Reed <greed@users.sourceforge.net> for reporting the
187 bug and providing a patch.
189 3. Added a compression option to dump, contributed by Jerry
190 Peters <gapeters@worldnet.att.net>.
192 WARNING: the compression works only on files/pipes and
193 drives supporting variable block size.
195 WARNING2: tapes created using the compression feature are
196 incompatible with the standard dump tape format, and a
197 version of restore >= 0.4b22 will be needed for extraction.
199 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
200 architectures. Thanks to Paul Slootman <paul@debian.org> for
201 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
204 5. Many cleanups (CPP defines, const char warnings, check of
205 ext2fs COMPAT flags, time_t cleanups, added libext2 version
206 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
208 6. Made --prefix option work in configure. All the install path
209 are now based on the configure parameters.
211 7. Added the Quick File Access mode in dump/restore, contributed
212 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
213 a file tape position for each inode, and this file is used by
214 restore (if called with parameter Q and the filename)
215 to directly position the tape at the file restore is currently
216 working on. This saves hours when restoring single files from
217 large backups, saves the tapes and the drive's head. Use
218 --enable-qfa option of configure to compile in the QFA support.
220 8. Added the possibility to dump several files and directories
221 in a single invocation of dump. Thanks to Uwe Gohlke
222 <uwe@ugsoft.de> for implementing this option.
224 9. Fixed the dumping and restoring of files with holes
225 (files ending with holes didn't get dumped or restored
228 10. Fixed a socket descriptor leak in dump, which leaved opened
229 3 file descriptors per dump process (and there is one dump
232 11. Fixed dump large file system support, by explicit use of
233 open64/lseek64/etc functions (explicit use needed because
234 e2fsprogs libraries don't behave well when compiled with
235 FILE_OFFSET_BITS=64).
237 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
238 ======================================================================
240 1. Fixed some bugs in the dump -w|-W logic introduced by
241 the previous version. Thanks to Andreas Dilger
242 <adilger@turbolinux.com> for his help on this one.
244 2. Fixed again a compilation problem when using old e2fs
245 headers (filesystem label related). Thanks to many users
246 who reported this stupid error.
248 3. Fixed a build problem on old lib5 based systems dealing with
249 _PATH_MNTTAB being undefined. Thanks to John Adams
250 <johna@onevista.com> for reporting the bug.
252 4. Improved the error detection in restore when creating
253 the temporary files in TMPDIR. Now restore will corectly
254 report a 'no space left on device' error instead of
255 strange errors which could imply an archive corruption.
256 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
257 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
259 5. Added the throughput information in dump's progression
260 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
263 6. Use libext2fs's inode scanning routines, which are particularly
264 robust when dealing with errors (especially when having some
265 bad blocks on the filesystem). Many thanks to Colin
266 <colin@horizon.com> for reporting the bug, investigating
267 the issues, finding a workaround, writing the patch and
268 fully testing it... (of course, if this patch will break
269 anything, he is to blame for :-)).
271 7. Made dump and restore LFS aware. Dump can dump a filesystem
272 containing large files, generate a large file on output and
273 restore can restore them. This feature is not enabled by
274 default, use --enable-largefile option of configure to enable
275 it (you need to have a LFS aware glibc though). Thanks to
276 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
277 and to Theodore T'so <tytso@valinux.com> for his always
280 8. Made dump ask upon a tape write error if it should rewrite
281 the current volume (assume this is a bad tape) or if it should
282 assume an end-of-tape condition (useful for tape drives which
283 behaves badly at the end of the tape). Thanks to Andreas
284 Dilger <adilger@turbolinux.com> for the suggestion.
286 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
287 =======================================================================
289 1. Fixed a small compilation problem due to a change
290 in the definintion of the struct sigaction in
291 glibc 2.0 / libc5. Thanks to Gunther Schlegel
292 <schlegel@riege.de> for reporting the bug and to
293 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
296 2. Modified the label and uuid analysis in order to be
297 self-contained (independant of kernel/libc headers). This
298 should fix the compile with older kernel/libc headers and
299 will preserve the functionality. Thanks to Bernhard Erdmann
300 <bernhard.erdmann@gmx.de> for reporting the bug.
302 3. The 'exclude inode' option, if applied to a directory
303 inode, excludes now correctly all the leaves of this
304 directory. Thanks to John R. Dennison
305 <gerdesas@users.sourceforge.net> for reporting the bug.
307 4. Fixed the '-e' option to disable the possibility
308 to exclude the root inode (which causes the dumps to
309 be unreadable by restore). Prevented array overflow
310 when multiple -e options are used.
312 5. Fixed dump to correctly interpret a filesystem argument
313 which is a mountpoint and it is not an absolute path
314 (as specified in the fstab). Thanks to Bernhard R. Erdmann
315 <be@berdmann.de> for reporting the bug.
317 6. Made dump able to backup files larger than 2 GB. Note that
318 dump still doesn't cope with files larger than 4 GB.
320 7. Restore the real uid and gid before invoking an external
321 RSH program (big hole when dump or restore is suid root!).
323 8. Get the values for the default tape device and the location
324 of fstab file from the system headers. Thanks to
325 Andreas Dilger <adilger@turbolinux.com> for the patch.
327 9. Made dump -w|-W report all recognized filesystems
328 present in either /etc/fstab or /etc/dumpdates, and present
329 the list in the same order as in fstab file. Thanks
330 to Andreas Dilger <adilger@turbolinux.com> for the patch.
332 10. Made dump's -a (automatic end of tape detection) the
333 default. Specifying one of -B, -c, -d or -s options will
334 override the EOM detection. Thanks to Andreas Dilger
335 <adilger@turbolinux.com> for the patch.
337 11. Save the ext2 filesystem label into the dump volume label.
338 Specifying a volume label on the command line (-L option)
339 will override this feature. Thanks to Andreas Dilger
340 <adilger@turbolinux.com> for the patch.
342 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
343 =====================================================================
345 1. Fixed the signal handling in dump (which I broke in 0.4b17)
346 which was causing several strange dump failures (dump
347 hanged or segmentation faults etc).
349 2. Specified the default blocksize in the dump man page.
351 3. Changed two info messages of restore to be written on stdout
352 instead of stderr in order to leave stderr only for errors
353 or warnings. Thanks to Stephen Atwell
354 <satwell@urbana.css.mot.com> for the suggestion.
356 4. Corrected an off by one calculation which prevented
357 dumping a directory having a 1 character name length.
358 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
359 for reporting the bug.
361 5. Reinforce dump in order to allow only level 0 and no
362 -u option when dumping a subdirectory, and document
363 this in the man page. Thanks to Bernhard Erdmann
364 <bernhard.erdmann@gmx.de> for reporting the bug.
366 6. Fixed a small harmless bug in restore which caused
367 in some conditions hard links to be restored several
368 times (therefore generation some warning messages).
369 Thanks to Kenneth Porter <shiva@well.com> for
372 7. Updated the RPM spec file to the latest RedHat version,
373 providing FHS packaging and other cosmetic changes.
374 You will need RPM version >= 3.0.5 in order to build the RPM.
376 8. Updated the configure script to check for libtermcap
377 before checking for libreadline, because we need this
378 library in order to compile the readline support.
380 9. Made dump understand the LABEL= and UUID= notation
381 both in /etc/fstab and on the command line. Note that
382 you will need the /proc filesystem in order to use
383 these notations. Thanks to Erik Troan <ewt@redhat.com>
384 for providing the patch.
386 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
387 ===================================================================
389 1. Fixed a potential buffer overflow in restore. Thanks
390 to Stan Bubrouski <satan@fastdial.net> for reporting
393 2. Fixed a readline-related bug which prevented
394 'cat DUMPFILE | restore -i -f -' from working. Thanks
395 to Charles Karney <karney@users.sourceforge.net>
398 3. Changed a few "panic" into "exit", causing restore to
399 be more stable against some attacks (like the last one
400 reported on Bugtraq, although the last version of restore
401 was not vulnerable - just dumped core). Thanks to
402 Andreas Hasenack <andreas@conectiva.com.br> for reporting
405 4. Removed the suid-root bit on dump and restore in the
406 default build (and generated RPMs). It should be safer
407 now. Users who need the suid-root capabilities in order
408 to make network backups should read first the man page
409 of dump and enable the suid bit manually.
411 5. Added -ltermcap to the compile parameters for restore
412 when using readline support, in order to make the compile
413 process work on some systems (SuSE ?). Thanks to
414 Patrik Schindler <poc@pocnet.net> for reporting the bug.
416 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
417 ====================================================================
419 1. The -F script is called now *only* at the end of a tape,
420 not at the end of the dump. Thanks to Christian Haul
421 <haul@informatik.tu-darmstadt.de> for the bug report.
423 Normally, the device name and the current volume number
424 are passed on the command line of the script. However,
425 if you want the old style script (with no arguments
426 passed on the command line), you can enable it in
427 configure with the --enable-oldstylefscript.
429 2. Use posix signal handling to preserve dump functionality
430 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
433 3. Made the exit code of restore in case of a 'C'ompare
434 command reflect the result of the compare. An exit status
435 of 0 means the dump archive is correct, 1 means tape errors,
436 2 means that some files were modified. Thanks to Kenneth Porter
437 <shiva@well.com> for the suggestion.
439 4. Made (finally) quotes work in interactive restore.
441 5. Small fixes in order to allow dump to compile with a
442 really old e2fsprogs version. Thanks to Ian Zimmerman
443 <itz@speakeasy.org> for the bug report.
445 6. Add GNU readline capabilities to interactive restore.
446 Use configure's --enable-readline flag to enable this feature.
447 Thanks to Patrik Schindler <poc@pocnet.net> for the
450 7. Do the compare on the fly in restore 'C' mode (this will
451 allow not to exhaust the available /tmp space when
452 ccmparing large files). Thanks to Kenneth Porter
453 <shiva@well.com> for the suggestion.
455 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
456 ====================================================================
458 1. Fixed some several duplicate 'DUMP: DUMP:' in the
461 2. Corrected the estimation of blocks to dump. Note that
462 this is still wrong for large directory inodes, due
463 to the size of a BSD directory entry on the tape
464 (estimation is based on the size of the ext2 directory,
465 which is smaller because it doesn't contain the null
466 character at the end).
468 3. Made dump report the total number of MB written to
469 tape at the end of run. Thanks to W. Reilly Cooley
470 <wcooley@nakedape.cc> for the patch.
472 4. Added the -X option to restore giving the possibility
473 to read the names of the files to be extracted/listed
474 from a text file (in addition of the command line).
475 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
478 5. Added the device name and the current volume number
479 as arguments to the end of tape script (dump -F option).
481 6. Made the multi-volume dump work again (after having
482 broken it in 0.4b15).
484 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
485 ===================================================================
487 1. Added a prompt command in interactive restore mode. Thanks
488 to Andreas Dilger <adilger@home.com> for the patch.
490 2. Fixed a buffer overflow problem in dump (caused by
491 not checking the size of the filesystem parameter).
492 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
493 reporting this on Bugtraq (and to several dump users
494 who forwarded me his mail).
496 3. Added the '-F script' option to dump in order to
497 launch a script at the end of each tape (to be used
498 with a tape changer, or to notify the sysadmin by
501 4. Fixed a bug in restore compare code caused by the changes
504 5. Fixed the treatment of options using the old BSD syntax
505 in both dump and restore.
507 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
508 =======================================================================
510 1. Fixed a bug in dump which may have caused invalid deleted
511 directories to be dumped out if they were deleted after the
512 mapping pass. This could occure on active filesystem and lead
513 to heap corruption (causing dump malfunction in many possible ways).
514 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
515 bug and submitting the patch.
517 2. Fixed the handling of the filesystem parameter in dump. A
518 '/mnt/disk' parameter caused the disk contents to be dumped,
519 but a '/mnt/disk/' parameter caused the mountpoint directory
520 to be dumped (generally an empty directory).
522 3. Improved the output of dump in order to tell which directory
523 it is currently dumping (when dumping a subtree).
525 4. Added the '-e' exclude inode option to dump. Thanks to
526 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
528 5. Added a REPORTING-BUGS file in order to provide a guide
529 on how to correctly report a bug in dump/restore.
531 6. Fixed a restore problem when restoring a hard link to an inode
532 having the immutable or append-only attribute set. Thanks to
533 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
535 7. Fixed a compatibility problem between dumps made on little
536 endian machines (the format was correct) and big endian
537 machines (the format was incorrect). This fix break the
538 compatibility with the older dumps made on big endian
539 machines (sparc, m86k, ppc etc). For the first time in
540 linux dump's history (I believe), the dumps made by *BSD,
541 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
542 so interchangeable. Thanks to Rob Cermak
543 <cermak@ahab.rutgers.edu> for submitting the bug and
544 helping me test the fix.
546 8. Fixed the way dump reports the remaining percent/time, if
547 the number of blocks actually dumped exceeds the estimated
548 number of blocks. Thanks to Jean-Paul van der Jagt
549 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
551 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
552 ======================================================================
554 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
555 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
557 2. Removed the requirement to build the RPM as root from the
558 spec file. Thanks to Christian Weisgerber
559 <naddy@mips.rhein-neckar.de> for submitting this.
561 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
562 causing dump to block when dumping really huge filesystems.
563 Many thanks to Patrik Schindler <poc@pocnet.net> for
564 helping me find this bug.
566 4. Fixed the treatment for an interrupt signal when dump access
567 the remote tape through RSH. Thanks to Christian Weisgerber
568 <naddy@mips.rhein-neckar.de> for providing the patch.
570 5. Fixed a bug which was causing dump/restore to display
571 garbage characters instead of the remote host name.
573 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
574 =====================================================================
576 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
577 <kukuk@suse.de> for submitting the patch.
579 2. Fix for the exit code when using the size estimate option of
580 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
581 submitting the patch.
583 3. Handle EINTR in atomical reads/writes in dump, which was causing
584 dump to fail on some systems. Thanks to Eric Jergensen
585 <eric@dvns.com> for reporting the bug and submitting the patch.
587 4. Handle more than 16 characters for the device names in dumpdates.
588 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
589 tracking down the problem and proposing the solution.
591 5. Fixed a bug in dump which prevented the creation of the
592 dumpdates file when doing a 0-level dump without already
593 having a dumpdates file. Thanks to Patrik Schindler
594 <poc@pocnet.net> for reporting the bug.
596 6. Changed the way dump 'S' flag reports the size estimate
597 from number of blocks into bytes (making it compatible
598 with the Solaris version, and simplifying things for
599 amanda users). Thanks to Jason L Tibbitts III
600 <tibbs@math.uh.edu> for reporting the bug.
602 7. Fixed a compatibility problem in linux/alpha dump tape format.
603 Now the linux/alpha dump are (again) compatible with the
604 other dump formats. But this breaks compatibility with
605 older dumps made on alpha. Thanks to Mike Tibor
606 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
608 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
609 ======================================================================
611 1. Added a '--enable-kerberos' to configure.
613 2. Added a 'S' option to dump which determines the amount of space
614 that is needed to perform the dump without actually doing it, similar
615 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
616 <cermak@ahab.rutgers.edu>.
618 3. Added a 'M' multi-volume option to dump and restore which enables
619 dumping to multiple files (useful when dumping to an ext2
620 partition to make several dump files in order to bypass the 2GB
621 file size limitation). The 'f' argument is treated as a prefix and
622 the output files will be named <prefix>001, <prefix>002 etc. With
623 the 'M' flag, restore automatically selects the right file without
624 asking to enter a new tape each time.
626 4. Fixed a memory leak which was causing dump to grow very big
627 (270MB when dumping a 10GB filesystem...). Thanks to Jason
628 Fearon <jasonf@netrider.org.au> for reporting the bug.
630 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
631 ======================================================================
633 1. Make configure test if the system glob routines support
634 extended syntax (ALTDIRFUNC). If not, use the internal glob
635 routines instead of system ones. Thanks to Bernhard Sadlowski
636 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
637 and helping me resolve this and other minor libc5 compiling
640 2. Fix a problem when dumping a ext2fs with the 'filetype'
641 feature enabled. Thanks to Patrick J. LoPresti
642 <patl@cag.lcs.mit.edu> for reporting the bug and to
643 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
645 3. Made the nodump flag work on directories. A directory which
646 has the nodump flag gets never dumped, regardless of its
649 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
650 which allows dump on an active ext3 filesystem. However, this
651 is a "quick and dirty" patch which enables backup of an ext3
652 filesystem through the ext2 compatibility (by ignoring the
653 NEEDS_RECOVERY bit). The journal file is not recognized and
654 it is dumped (it should not).
656 5. Test the superblock compatibility flags when dumping, in order
657 to be sure that we know how to deal with specific features.
659 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
660 ====================================================================
662 1. Use lchown instead of chown, fixing a possible security problem
663 when restoring symlinks (a malicious user could use this
664 to deliberately corrupt the ownership of important system files).
665 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
666 this and providing the patch.
668 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
669 ====================================================================
671 1. Put dump sources under CVS, added Id tags in all files so
672 one can use 'ident' on binary files.
674 2. Added the dump/restore version in the usage text so one can
675 easily verify the version he is using.
677 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
678 a va_start/va_end problem on linux-ppc (always call va_start
679 va_end in pairs each time we use a vararg function).
681 4. Added again the DT_* constants because old libc does not
682 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
683 for submitting the bug report.
685 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
686 this should enable dumping big (huge) filesystems.
688 6. Added the RSH environment variable in order to be able to
689 use a rsh replacement like ssh when doing remote backups (and
690 bypass the security limitations of rcmd). Now you can do remote
691 backups without being root (or making dump setuid root).
693 7. Modified again the way dumpdates works. For incremental dumps,
694 we need to read dumpdates even if we are not using 'u' option.
695 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
698 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
699 ===================================================================
701 1. Removed the 'k' flag from the restore 'about' text if kerberos
704 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
705 to them (fsetflags takes a char*, setflags an open fd!).
707 3. (f)setflags is called only if the flags aren't empty. If the
708 file is a special file, a warning is printed, because changing
709 flags implies opening the device. Normally, a special file
710 should not have any flag... (Debian bug #29775, patch provided
711 by Abhijit Dasgupta <abhijit@ans.net>).
713 4. Made possible to dump a file system not mentioned in /etc/fstab.
714 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
716 5. Changed the default behaviour to not create dumpdates
717 unless 'u' option is specified. Removed the old "debian-patch"
718 which provided the same thing. (Debian bug #38136, #33818).
720 6. Removed all those dump*announce, since they were getting old...
722 7. Added warning messages if dumpdates does not exist and
723 when an inferior level dump does not exist (except for a level 0
726 8. Debugged the glob calls in interactive mode: restore used a
727 dirent struct which was different from the /usr/include/dirent.h
728 one (this used to work, is it a glibc2 change?), so none of the
729 compat glob (which used /usr/include/dirent.h) or the system glob
730 worked. Restore use now the system dirent (and the system
731 DT_* constants), which are compatible with BSD ones.
733 9. Added a configure flag (--with-dumpdatespath) to specify
734 the location of dumpdates file. By default, it is
737 10. Added the "AUTHOR" and "AVAILABILITY" sections and
738 included the current date/version in man pages.
740 11. Corrected the estimation of remaining time when
741 the operator doesn't change the tapes quickly enough. This
742 was an old bug, I thought I corrected it, and discovered
743 that in fact it was corrected in two different places, so
744 the results canceled each other...
746 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
747 ===================================================================
749 1. Integrated multiple patches from RedHat, Debian and SuSE:
751 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
753 - several fixes in the man pages.
754 - update the default tape device to /dev/st0.
755 - many updates for Linux Alpha (byte ordering, size_t etc).
757 - use environment variable for TMPDIR (instead of /tmp).
758 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
759 - workaround egcs bug (RedHat bugs #4281 and #2989).
760 - wire $(OPT) throughout Makefile's.
762 2. Upgrade the dump revision to 1, making possible to dump filesystems
763 made with e2fsprogs-1.15 or newer. Nothing seems to break...
765 3. Fix some compile warnings, prototype all functions.
767 4. Use glibc err/glob instead of internal compatibility
768 routines (only if available).
770 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
771 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
773 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
774 ======================================================================
776 1. Integrated the changes from FreeBSD-3.1-RELEASE
777 (mostly bug fixes, buffer overruns, dump has now an "automatic
778 tape length calculation" flag, dump/restore can use kerberos now
779 (this is NOT tested), use environment variables for TAPE and
782 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
784 3. Corrected a bug in restore when using 'C' option with multi-volumes
785 tapes (files splited accros two tapes give "size changed" errors
788 4. Corrected the long standing bug when dumping multiple tapes.
789 This works for me, needs further testing.
791 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
792 ====================================================================
794 1. Dump now runs correctly on kernels 2.1.x
795 Fix made by Gerald Peters <gapeters@worldnet.att.net>
797 Changes between versions 0.4b2 and 0.4b3
798 ========================================
800 1. Use realpath() if available
804 Changes between versions 0.4b1 and 0.4b2
805 ========================================
807 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
810 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
812 3. Fixed configure.in to correctly handle --enable-rmt
814 Changes between versions 0.3 and 0.4b1
815 ======================================
817 1. Integrated the changes from 4.4BSD-Lite2
819 2. Integrated the patches from Debian and Red Hat
821 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
823 4. Changed dump to use the Ext2fs library to get block addresses. This
824 should solve the endianness problem on SparcLinux.
826 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
827 distribution's one) to use autoconf
829 6. Fixed a few minor bugs
831 Changes between versions 0.2e and 0.2f
832 ======================================
834 1. Added the creation of named pipes (fifos) in restore.
836 2. Added the -N flag in the restore manual page.
838 3. Added the file kernel.patch which contains the llseek() optimization
839 patch for 1.2.x kernels.
841 4. Fixed a bug in the restoration of symbolic links: owner and group were
844 5. Integrated some changes from FreeBSD 2.2.
846 6. Added a call to ftruncate() after restoring each file to restore
847 correctly files ending by a hole.
849 Changes between versions 0.2d and 0.2e
850 ======================================
852 1. Fixed a bug in the "set owner/mode" process. Every file was restored
853 with owner = root (0) and group = root/wheel/whatever (0).
855 Changes between versions 0.2c and 0.2d
856 ======================================
858 1. Dump is now able to backup 2GB+ filesystems.
860 2. Dump and restore can now be linked as static binaries.
862 Changes between versions 0.2b and 0.2c
863 ======================================
865 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
867 Changes between versions 0.2a and 0.2b
868 ======================================
870 1. Really fixed the bug that I should have corrected in 0.2a.
872 2. Enabled optimization again.
874 Changes between versions 0.2 and 0.2a
875 =====================================
877 1. Disabled the optimization during compilation.
879 Changes between versions 0.1 and 0.2
880 ====================================
882 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
883 the fs_type field (actually, I modified the file fstab.c to make it
884 use the mntent functions).
886 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
887 made by dump should be readable by the BSD restore and backups made
888 by the BSD dump should be readable by restore. Unfortunately, this
889 means that the dump format has changed between version 0.1 and version
892 3. Dump is now able to backup a subtree, it is no longer limited to whole
893 filesystems like the BSD version.
895 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
898 Changes between versions 0.0 and 0.1
899 ====================================
901 1. Now create links rdump and rrestore during the `make install' step.
903 2. Linux port specific bugs added to the manual pages
905 3. Incorrect estimation of the number of tapes blocks fixed when doing
908 4. Better ls-like format in restore in interactive mode.