1 $Id: CHANGES,v 1.86 2000/11/10 09:35:07 stelian Exp $
3 Changes between versions 0.4b19 and 0.4b20 (released ???????????????)
4 =====================================================================
6 1. Fixed a small compilation problem due to a change
7 in the definintion of the struct sigaction in
8 glibc 2.0 / libc5. Thanks to Gunther Schlegel
9 <schlegel@riege.de> for reporting the bug and to
10 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
13 2. Modified the label and uuid analysis in order to be
14 self-contained (independant of kernel/libc headers). This
15 should fix the compile with older kernel/libc headers and
16 will preserve the functionality. Thanks to Bernhard Erdmann
17 <bernhard.erdmann@gmx.de> for reporting the bug.
19 3. The 'exclude inode' option, if applied to a directory
20 inode, excludes now correctly all the leaves of this
21 directory. Thanks to John R. Dennison
22 <gerdesas@users.sourceforge.net> for reporting the bug.
24 4. Fixed the '-e' option to disable the possibility
25 to exclude the root inode (which causes the dumps to
26 be unreadable by restore). Prevented array overflow
27 when multiple -e options are used.
29 5. Fixed dump to correctly interpret a filesystem argument
30 which is a mountpoint and it is not an absolute path
31 (as specified in the fstab). Thanks to Bernhard R. Erdmann
32 <be@berdmann.de> for reporting the bug.
34 6. Made dump able to backup files larger than 2 GB. Note that
35 dump still doesn't cope with files larger than 4 GB.
37 7. Restore the real uid and gid before invoking an external
38 RSH program (big hole when dump or restore is suid root!).
40 8. Get the values for the default tape device and the location
41 of fstab file from the system headers. Thanks to
42 Andreas Dilger <adilger@turbolinux.com> for the patch.
44 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
45 =====================================================================
47 1. Fixed the signal handling in dump (which I broke in 0.4b17)
48 which was causing several strange dump failures (dump
49 hanged or segmentation faults etc).
51 2. Specified the default blocksize in the dump man page.
53 3. Changed two info messages of restore to be written on stdout
54 instead of stderr in order to leave stderr only for errors
55 or warnings. Thanks to Stephen Atwell
56 <satwell@urbana.css.mot.com> for the suggestion.
58 4. Corrected an off by one calculation which prevented
59 dumping a directory having a 1 character name length.
60 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
61 for reporting the bug.
63 5. Reinforce dump in order to allow only level 0 and no
64 -u option when dumping a subdirectory, and document
65 this in the man page. Thanks to Bernhard Erdmann
66 <bernhard.erdmann@gmx.de> for reporting the bug.
68 6. Fixed a small harmless bug in restore which caused
69 in some conditions hard links to be restored several
70 times (therefore generation some warning messages).
71 Thanks to Kenneth Porter <shiva@well.com> for
74 7. Updated the RPM spec file to the latest RedHat version,
75 providing FHS packaging and other cosmetic changes.
76 You will need RPM version >= 3.0.5 in order to build the RPM.
78 8. Updated the configure script to check for libtermcap
79 before checking for libreadline, because we need this
80 library in order to compile the readline support.
82 9. Made dump understand the LABEL= and UUID= notation
83 both in /etc/fstab and on the command line. Note that
84 you will need the /proc filesystem in order to use
85 these notations. Thanks to Erik Troan <ewt@redhat.com>
86 for providing the patch.
88 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
89 ===================================================================
91 1. Fixed a potential buffer overflow in restore. Thanks
92 to Stan Bubrouski <satan@fastdial.net> for reporting
95 2. Fixed a readline-related bug which prevented
96 'cat DUMPFILE | restore -i -f -' from working. Thanks
97 to Charles Karney <karney@users.sourceforge.net>
100 3. Changed a few "panic" into "exit", causing restore to
101 be more stable against some attacks (like the last one
102 reported on Bugtraq, although the last version of restore
103 was not vulnerable - just dumped core). Thanks to
104 Andreas Hasenack <andreas@conectiva.com.br> for reporting
107 4. Removed the suid-root bit on dump and restore in the
108 default build (and generated RPMs). It should be safer
109 now. Users who need the suid-root capabilities in order
110 to make network backups should read first the man page
111 of dump and enable the suid bit manually.
113 5. Added -ltermcap to the compile parameters for restore
114 when using readline support, in order to make the compile
115 process work on some systems (SuSE ?). Thanks to
116 Patrik Schindler <poc@pocnet.net> for reporting the bug.
118 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
119 ====================================================================
121 1. The -F script is called now *only* at the end of a tape,
122 not at the end of the dump. Thanks to Christian Haul
123 <haul@informatik.tu-darmstadt.de> for the bug report.
125 Normally, the device name and the current volume number
126 are passed on the command line of the script. However,
127 if you want the old style script (with no arguments
128 passed on the command line), you can enable it in
129 configure with the --enable-oldstylefscript.
131 2. Use posix signal handling to preserve dump functionality
132 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
135 3. Made the exit code of restore in case of a 'C'ompare
136 command reflect the result of the compare. An exit status
137 of 0 means the dump archive is correct, 1 means tape errors,
138 2 means that some files were modified. Thanks to Kenneth Porter
139 <shiva@well.com> for the suggestion.
141 4. Made (finally) quotes work in interactive restore.
143 5. Small fixes in order to allow dump to compile with a
144 really old e2fsprogs version. Thanks to Ian Zimmerman
145 <itz@speakeasy.org> for the bug report.
147 6. Add GNU readline capabilities to interactive restore.
148 Use configure's --enable-readline flag to enable this feature.
149 Thanks to Patrik Schindler <poc@pocnet.net> for the
152 7. Do the compare on the fly in restore 'C' mode (this will
153 allow not to exhaust the available /tmp space when
154 ccmparing large files). Thanks to Kenneth Porter
155 <shiva@well.com> for the suggestion.
157 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
158 ====================================================================
160 1. Fixed some several duplicate 'DUMP: DUMP:' in the
163 2. Corrected the estimation of blocks to dump. Note that
164 this is still wrong for large directory inodes, due
165 to the size of a BSD directory entry on the tape
166 (estimation is based on the size of the ext2 directory,
167 which is smaller because it doesn't contain the null
168 character at the end).
170 3. Made dump report the total number of MB written to
171 tape at the end of run. Thanks to W. Reilly Cooley
172 <wcooley@nakedape.cc> for the patch.
174 4. Added the -X option to restore giving the possibility
175 to read the names of the files to be extracted/listed
176 from a text file (in addition of the command line).
177 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
180 5. Added the device name and the current volume number
181 as arguments to the end of tape script (dump -F option).
183 6. Made the multi-volume dump work again (after having
184 broken it in 0.4b15).
186 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
187 ===================================================================
189 1. Added a prompt command in interactive restore mode. Thanks
190 to Andreas Dilger <adilger@home.com> for the patch.
192 2. Fixed a buffer overflow problem in dump (caused by
193 not checking the size of the filesystem parameter).
194 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
195 reporting this on Bugtraq (and to several dump users
196 who forwarded me his mail).
198 3. Added the '-F script' option to dump in order to
199 launch a script at the end of each tape (to be used
200 with a tape changer, or to notify the sysadmin by
203 4. Fixed a bug in restore compare code caused by the changes
206 5. Fixed the treatment of options using the old BSD syntax
207 in both dump and restore.
209 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
210 =======================================================================
212 1. Fixed a bug in dump which may have caused invalid deleted
213 directories to be dumped out if they were deleted after the
214 mapping pass. This could occure on active filesystem and lead
215 to heap corruption (causing dump malfunction in many possible ways).
216 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
217 bug and submitting the patch.
219 2. Fixed the handling of the filesystem parameter in dump. A
220 '/mnt/disk' parameter caused the disk contents to be dumped,
221 but a '/mnt/disk/' parameter caused the mountpoint directory
222 to be dumped (generally an empty directory).
224 3. Improved the output of dump in order to tell which directory
225 it is currently dumping (when dumping a subtree).
227 4. Added the '-e' exclude inode option to dump. Thanks to
228 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
230 5. Added a REPORTING-BUGS file in order to provide a guide
231 on how to correctly report a bug in dump/restore.
233 6. Fixed a restore problem when restoring a hard link to an inode
234 having the immutable or append-only attribute set. Thanks to
235 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
237 7. Fixed a compatibility problem between dumps made on little
238 endian machines (the format was correct) and big endian
239 machines (the format was incorrect). This fix break the
240 compatibility with the older dumps made on big endian
241 machines (sparc, m86k, ppc etc). For the first time in
242 linux dump's history (I believe), the dumps made by *BSD,
243 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
244 so interchangeable. Thanks to Rob Cermak
245 <cermak@ahab.rutgers.edu> for submitting the bug and
246 helping me test the fix.
248 8. Fixed the way dump reports the remaining percent/time, if
249 the number of blocks actually dumped exceeds the estimated
250 number of blocks. Thanks to Jean-Paul van der Jagt
251 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
253 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
254 ======================================================================
256 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
257 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
259 2. Removed the requirement to build the RPM as root from the
260 spec file. Thanks to Christian Weisgerber
261 <naddy@mips.rhein-neckar.de> for submitting this.
263 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
264 causing dump to block when dumping really huge filesystems.
265 Many thanks to Patrik Schindler <poc@pocnet.net> for
266 helping me find this bug.
268 4. Fixed the treatment for an interrupt signal when dump access
269 the remote tape through RSH. Thanks to Christian Weisgerber
270 <naddy@mips.rhein-neckar.de> for providing the patch.
272 5. Fixed a bug which was causing dump/restore to display
273 garbage characters instead of the remote host name.
275 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
276 =====================================================================
278 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
279 <kukuk@suse.de> for submitting the patch.
281 2. Fix for the exit code when using the size estimate option of
282 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
283 submitting the patch.
285 3. Handle EINTR in atomical reads/writes in dump, which was causing
286 dump to fail on some systems. Thanks to Eric Jergensen
287 <eric@dvns.com> for reporting the bug and submitting the patch.
289 4. Handle more than 16 characters for the device names in dumpdates.
290 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
291 tracking down the problem and proposing the solution.
293 5. Fixed a bug in dump which prevented the creation of the
294 dumpdates file when doing a 0-level dump without already
295 having a dumpdates file. Thanks to Patrik Schindler
296 <poc@pocnet.net> for reporting the bug.
298 6. Changed the way dump 'S' flag reports the size estimate
299 from number of blocks into bytes (making it compatible
300 with the Solaris version, and simplifying things for
301 amanda users). Thanks to Jason L Tibbitts III
302 <tibbs@math.uh.edu> for reporting the bug.
304 7. Fixed a compatibility problem in linux/alpha dump tape format.
305 Now the linux/alpha dump are (again) compatible with the
306 other dump formats. But this breaks compatibility with
307 older dumps made on alpha. Thanks to Mike Tibor
308 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
310 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
311 ======================================================================
313 1. Added a '--enable-kerberos' to configure.
315 2. Added a 'S' option to dump which determines the amount of space
316 that is needed to perform the dump without actually doing it, similar
317 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
318 <cermak@ahab.rutgers.edu>.
320 3. Added a 'M' multi-volume option to dump and restore which enables
321 dumping to multiple files (useful when dumping to an ext2
322 partition to make several dump files in order to bypass the 2GB
323 file size limitation). The 'f' argument is treated as a prefix and
324 the output files will be named <prefix>001, <prefix>002 etc. With
325 the 'M' flag, restore automatically selects the right file without
326 asking to enter a new tape each time.
328 4. Fixed a memory leak which was causing dump to grow very big
329 (270MB when dumping a 10GB filesystem...). Thanks to Jason
330 Fearon <jasonf@netrider.org.au> for reporting the bug.
332 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
333 ======================================================================
335 1. Make configure test if the system glob routines support
336 extended syntax (ALTDIRFUNC). If not, use the internal glob
337 routines instead of system ones. Thanks to Bernhard Sadlowski
338 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
339 and helping me resolve this and other minor libc5 compiling
342 2. Fix a problem when dumping a ext2fs with the 'filetype'
343 feature enabled. Thanks to Patrick J. LoPresti
344 <patl@cag.lcs.mit.edu> for reporting the bug and to
345 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
347 3. Made the nodump flag work on directories. A directory which
348 has the nodump flag gets never dumped, regardless of its
351 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
352 which allows dump on an active ext3 filesystem. However, this
353 is a "quick and dirty" patch which enables backup of an ext3
354 filesystem through the ext2 compatibility (by ignoring the
355 NEEDS_RECOVERY bit). The journal file is not recognized and
356 it is dumped (it should not).
358 5. Test the superblock compatibility flags when dumping, in order
359 to be sure that we know how to deal with specific features.
361 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
362 ====================================================================
364 1. Use lchown instead of chown, fixing a possible security problem
365 when restoring symlinks (a malicious user could use this
366 to deliberately corrupt the ownership of important system files).
367 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
368 this and providing the patch.
370 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
371 ====================================================================
373 1. Put dump sources under CVS, added Id tags in all files so
374 one can use 'ident' on binary files.
376 2. Added the dump/restore version in the usage text so one can
377 easily verify the version he is using.
379 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
380 a va_start/va_end problem on linux-ppc (always call va_start
381 va_end in pairs each time we use a vararg function).
383 4. Added again the DT_* constants because old libc does not
384 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
385 for submitting the bug report.
387 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
388 this should enable dumping big (huge) filesystems.
390 6. Added the RSH environment variable in order to be able to
391 use a rsh replacement like ssh when doing remote backups (and
392 bypass the security limitations of rcmd). Now you can do remote
393 backups without being root (or making dump setuid root).
395 7. Modified again the way dumpdates works. For incremental dumps,
396 we need to read dumpdates even if we are not using 'u' option.
397 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
400 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
401 ===================================================================
403 1. Removed the 'k' flag from the restore 'about' text if kerberos
406 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
407 to them (fsetflags takes a char*, setflags an open fd!).
409 3. (f)setflags is called only if the flags aren't empty. If the
410 file is a special file, a warning is printed, because changing
411 flags implies opening the device. Normally, a special file
412 should not have any flag... (Debian bug #29775, patch provided
413 by Abhijit Dasgupta <abhijit@ans.net>).
415 4. Made possible to dump a file system not mentioned in /etc/fstab.
416 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
418 5. Changed the default behaviour to not create dumpdates
419 unless 'u' option is specified. Removed the old "debian-patch"
420 which provided the same thing. (Debian bug #38136, #33818).
422 6. Removed all those dump*announce, since they were getting old...
424 7. Added warning messages if dumpdates does not exist and
425 when an inferior level dump does not exist (except for a level 0
428 8. Debugged the glob calls in interactive mode: restore used a
429 dirent struct which was different from the /usr/include/dirent.h
430 one (this used to work, is it a glibc2 change?), so none of the
431 compat glob (which used /usr/include/dirent.h) or the system glob
432 worked. Restore use now the system dirent (and the system
433 DT_* constants), which are compatible with BSD ones.
435 9. Added a configure flag (--with-dumpdatespath) to specify
436 the location of dumpdates file. By default, it is
439 10. Added the "AUTHOR" and "AVAILABILITY" sections and
440 included the current date/version in man pages.
442 11. Corrected the estimation of remaining time when
443 the operator doesn't change the tapes quickly enough. This
444 was an old bug, I thought I corrected it, and discovered
445 that in fact it was corrected in two different places, so
446 the results canceled each other...
448 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
449 ===================================================================
451 1. Integrated multiple patches from RedHat, Debian and SuSE:
453 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
455 - several fixes in the man pages.
456 - update the default tape device to /dev/st0.
457 - many updates for Linux Alpha (byte ordering, size_t etc).
459 - use environment variable for TMPDIR (instead of /tmp).
460 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
461 - workaround egcs bug (RedHat bugs #4281 and #2989).
462 - wire $(OPT) throughout Makefile's.
464 2. Upgrade the dump revision to 1, making possible to dump filesystems
465 made with e2fsprogs-1.15 or newer. Nothing seems to break...
467 3. Fix some compile warnings, prototype all functions.
469 4. Use glibc err/glob instead of internal compatibility
470 routines (only if available).
472 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
473 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
475 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
476 ======================================================================
478 1. Integrated the changes from FreeBSD-3.1-RELEASE
479 (mostly bug fixes, buffer overruns, dump has now an "automatic
480 tape length calculation" flag, dump/restore can use kerberos now
481 (this is NOT tested), use environment variables for TAPE and
484 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
486 3. Corrected a bug in restore when using 'C' option with multi-volumes
487 tapes (files splited accros two tapes give "size changed" errors
490 4. Corrected the long standing bug when dumping multiple tapes.
491 This works for me, needs further testing.
493 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
494 ====================================================================
496 1. Dump now runs correctly on kernels 2.1.x
497 Fix made by Gerald Peters <gapeters@worldnet.att.net>
499 Changes between versions 0.4b2 and 0.4b3
500 ========================================
502 1. Use realpath() if available
506 Changes between versions 0.4b1 and 0.4b2
507 ========================================
509 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
512 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
514 3. Fixed configure.in to correctly handle --enable-rmt
516 Changes between versions 0.3 and 0.4b1
517 ======================================
519 1. Integrated the changes from 4.4BSD-Lite2
521 2. Integrated the patches from Debian and Red Hat
523 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
525 4. Changed dump to use the Ext2fs library to get block addresses. This
526 should solve the endianness problem on SparcLinux.
528 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
529 distribution's one) to use autoconf
531 6. Fixed a few minor bugs
533 Changes between versions 0.2e and 0.2f
534 ======================================
536 1. Added the creation of named pipes (fifos) in restore.
538 2. Added the -N flag in the restore manual page.
540 3. Added the file kernel.patch which contains the llseek() optimization
541 patch for 1.2.x kernels.
543 4. Fixed a bug in the restoration of symbolic links: owner and group were
546 5. Integrated some changes from FreeBSD 2.2.
548 6. Added a call to ftruncate() after restoring each file to restore
549 correctly files ending by a hole.
551 Changes between versions 0.2d and 0.2e
552 ======================================
554 1. Fixed a bug in the "set owner/mode" process. Every file was restored
555 with owner = root (0) and group = root/wheel/whatever (0).
557 Changes between versions 0.2c and 0.2d
558 ======================================
560 1. Dump is now able to backup 2GB+ filesystems.
562 2. Dump and restore can now be linked as static binaries.
564 Changes between versions 0.2b and 0.2c
565 ======================================
567 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
569 Changes between versions 0.2a and 0.2b
570 ======================================
572 1. Really fixed the bug that I should have corrected in 0.2a.
574 2. Enabled optimization again.
576 Changes between versions 0.2 and 0.2a
577 =====================================
579 1. Disabled the optimization during compilation.
581 Changes between versions 0.1 and 0.2
582 ====================================
584 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
585 the fs_type field (actually, I modified the file fstab.c to make it
586 use the mntent functions).
588 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
589 made by dump should be readable by the BSD restore and backups made
590 by the BSD dump should be readable by restore. Unfortunately, this
591 means that the dump format has changed between version 0.1 and version
594 3. Dump is now able to backup a subtree, it is no longer limited to whole
595 filesystems like the BSD version.
597 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
600 Changes between versions 0.0 and 0.1
601 ====================================
603 1. Now create links rdump and rrestore during the `make install' step.
605 2. Linux port specific bugs added to the manual pages
607 3. Incorrect estimation of the number of tapes blocks fixed when doing
610 4. Better ls-like format in restore in interactive mode.