1 $Id: CHANGES,v 1.76 2000/08/20 16:34:00 stelian Exp $
3 Changes between versions 0.4b17 and 0.4b19 (released ?????????????)
4 ===================================================================
6 1. Fixed the signal handling in dump (which I broke in 0.4b17)
7 which was causing several strange dump failures (dump
8 hanged or segmentation faults etc).
10 2. Specified the default blocksize in the dump man page.
12 3. Changed two info messages of restore to be written on stdout
13 instead of stderr in order to leave stderr only for errors
14 or warnings. Thanks to Stephen Atwell
15 <satwell@urbana.css.mot.com> for the suggestion.
17 4. Corrected an off by one calculation which prevented
18 dumping a directory having a 1 character name length.
19 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
20 for reporting the bug.
22 5. Reinforce dump in order to allow only level 0 and no
23 -u option when dumping a subdirectory, and document
24 this in the man page. Thanks to Bernhard Erdmann
25 <bernhard.erdmann@gmx.de> for reporting the bug.
27 6. Fixed a small harmless bug in restore which caused
28 in some conditions hard links to be restored several
29 times (therefore generation some warning messages).
30 Thanks to Kenneth Porter <shiva@well.com> for
33 7. Updated the RPM spec file to the latest RedHat version,
34 providing FHS packaging and other cosmetic changes.
36 8. Updated the configure script to check for libtermcap
37 before checking for libreadline, because we need this
38 library in order to compile the readline support.
40 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
41 ===================================================================
43 1. Fixed a potential buffer overflow in restore. Thanks
44 to Stan Bubrouski <satan@fastdial.net> for reporting
47 2. Fixed a readline-related bug which prevented
48 'cat DUMPFILE | restore -i -f -' from working. Thanks
49 to Charles Karney <karney@users.sourceforge.net>
52 3. Changed a few "panic" into "exit", causing restore to
53 be more stable against some attacks (like the last one
54 reported on Bugtraq, although the last version of restore
55 was not vulnerable - just dumped core). Thanks to
56 Andreas Hasenack <andreas@conectiva.com.br> for reporting
59 4. Removed the suid-root bit on dump and restore in the
60 default build (and generated RPMs). It should be safer
61 now. Users who need the suid-root capabilities in order
62 to make network backups should read first the man page
63 of dump and enable the suid bit manually.
65 5. Added -ltermcap to the compile parameters for restore
66 when using readline support, in order to make the compile
67 process work on some systems (SuSE ?). Thanks to
68 Patrik Schindler <poc@pocnet.net> for reporting the bug.
70 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
71 ====================================================================
73 1. The -F script is called now *only* at the end of a tape,
74 not at the end of the dump. Thanks to Christian Haul
75 <haul@informatik.tu-darmstadt.de> for the bug report.
77 Normally, the device name and the current volume number
78 are passed on the command line of the script. However,
79 if you want the old style script (with no arguments
80 passed on the command line), you can enable it in
81 configure with the --enable-oldstylefscript.
83 2. Use posix signal handling to preserve dump functionality
84 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
87 3. Made the exit code of restore in case of a 'C'ompare
88 command reflect the result of the compare. An exit status
89 of 0 means the dump archive is correct, 1 means tape errors,
90 2 means that some files were modified. Thanks to Kenneth Porter
91 <shiva@well.com> for the suggestion.
93 4. Made (finally) quotes work in interactive restore.
95 5. Small fixes in order to allow dump to compile with a
96 really old e2fsprogs version. Thanks to Ian Zimmerman
97 <itz@speakeasy.org> for the bug report.
99 6. Add GNU readline capabilities to interactive restore.
100 Use configure's --enable-readline flag to enable this feature.
101 Thanks to Patrik Schindler <poc@pocnet.net> for the
104 7. Do the compare on the fly in restore 'C' mode (this will
105 allow not to exhaust the available /tmp space when
106 ccmparing large files). Thanks to Kenneth Porter
107 <shiva@well.com> for the suggestion.
109 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
110 ====================================================================
112 1. Fixed some several duplicate 'DUMP: DUMP:' in the
115 2. Corrected the estimation of blocks to dump. Note that
116 this is still wrong for large directory inodes, due
117 to the size of a BSD directory entry on the tape
118 (estimation is based on the size of the ext2 directory,
119 which is smaller because it doesn't contain the null
120 character at the end).
122 3. Made dump report the total number of MB written to
123 tape at the end of run. Thanks to W. Reilly Cooley
124 <wcooley@nakedape.cc> for the patch.
126 4. Added the -X option to restore giving the possibility
127 to read the names of the files to be extracted/listed
128 from a text file (in addition of the command line).
129 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
132 5. Added the device name and the current volume number
133 as arguments to the end of tape script (dump -F option).
135 6. Made the multi-volume dump work again (after having
136 broken it in 0.4b15).
138 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
139 ===================================================================
141 1. Added a prompt command in interactive restore mode. Thanks
142 to Andreas Dilger <adilger@home.com> for the patch.
144 2. Fixed a buffer overflow problem in dump (caused by
145 not checking the size of the filesystem parameter).
146 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
147 reporting this on Bugtraq (and to several dump users
148 who forwarded me his mail).
150 3. Added the '-F script' option to dump in order to
151 launch a script at the end of each tape (to be used
152 with a tape changer, or to notify the sysadmin by
155 4. Fixed a bug in restore compare code caused by the changes
158 5. Fixed the treatment of options using the old BSD syntax
159 in both dump and restore.
161 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
162 =======================================================================
164 1. Fixed a bug in dump which may have caused invalid deleted
165 directories to be dumped out if they were deleted after the
166 mapping pass. This could occure on active filesystem and lead
167 to heap corruption (causing dump malfunction in many possible ways).
168 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
169 bug and submitting the patch.
171 2. Fixed the handling of the filesystem parameter in dump. A
172 '/mnt/disk' parameter caused the disk contents to be dumped,
173 but a '/mnt/disk/' parameter caused the mountpoint directory
174 to be dumped (generally an empty directory).
176 3. Improved the output of dump in order to tell which directory
177 it is currently dumping (when dumping a subtree).
179 4. Added the '-e' exclude inode option to dump. Thanks to
180 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
182 5. Added a REPORTING-BUGS file in order to provide a guide
183 on how to correctly report a bug in dump/restore.
185 6. Fixed a restore problem when restoring a hard link to an inode
186 having the immutable or append-only attribute set. Thanks to
187 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
189 7. Fixed a compatibility problem between dumps made on little
190 endian machines (the format was correct) and big endian
191 machines (the format was incorrect). This fix break the
192 compatibility with the older dumps made on big endian
193 machines (sparc, m86k, ppc etc). For the first time in
194 linux dump's history (I believe), the dumps made by *BSD,
195 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
196 so interchangeable. Thanks to Rob Cermak
197 <cermak@ahab.rutgers.edu> for submitting the bug and
198 helping me test the fix.
200 8. Fixed the way dump reports the remaining percent/time, if
201 the number of blocks actually dumped exceeds the estimated
202 number of blocks. Thanks to Jean-Paul van der Jagt
203 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
205 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
206 ======================================================================
208 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
209 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
211 2. Removed the requirement to build the RPM as root from the
212 spec file. Thanks to Christian Weisgerber
213 <naddy@mips.rhein-neckar.de> for submitting this.
215 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
216 causing dump to block when dumping really huge filesystems.
217 Many thanks to Patrik Schindler <poc@pocnet.net> for
218 helping me find this bug.
220 4. Fixed the treatment for an interrupt signal when dump access
221 the remote tape through RSH. Thanks to Christian Weisgerber
222 <naddy@mips.rhein-neckar.de> for providing the patch.
224 5. Fixed a bug which was causing dump/restore to display
225 garbage characters instead of the remote host name.
227 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
228 =====================================================================
230 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
231 <kukuk@suse.de> for submitting the patch.
233 2. Fix for the exit code when using the size estimate option of
234 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
235 submitting the patch.
237 3. Handle EINTR in atomical reads/writes in dump, which was causing
238 dump to fail on some systems. Thanks to Eric Jergensen
239 <eric@dvns.com> for reporting the bug and submitting the patch.
241 4. Handle more than 16 characters for the device names in dumpdates.
242 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
243 tracking down the problem and proposing the solution.
245 5. Fixed a bug in dump which prevented the creation of the
246 dumpdates file when doing a 0-level dump without already
247 having a dumpdates file. Thanks to Patrik Schindler
248 <poc@pocnet.net> for reporting the bug.
250 6. Changed the way dump 'S' flag reports the size estimate
251 from number of blocks into bytes (making it compatible
252 with the Solaris version, and simplifying things for
253 amanda users). Thanks to Jason L Tibbitts III
254 <tibbs@math.uh.edu> for reporting the bug.
256 7. Fixed a compatibility problem in linux/alpha dump tape format.
257 Now the linux/alpha dump are (again) compatible with the
258 other dump formats. But this breaks compatibility with
259 older dumps made on alpha. Thanks to Mike Tibor
260 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
262 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
263 ======================================================================
265 1. Added a '--enable-kerberos' to configure.
267 2. Added a 'S' option to dump which determines the amount of space
268 that is needed to perform the dump without actually doing it, similar
269 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
270 <cermak@ahab.rutgers.edu>.
272 3. Added a 'M' multi-volume option to dump and restore which enables
273 dumping to multiple files (useful when dumping to an ext2
274 partition to make several dump files in order to bypass the 2GB
275 file size limitation). The 'f' argument is treated as a prefix and
276 the output files will be named <prefix>001, <prefix>002 etc. With
277 the 'M' flag, restore automatically selects the right file without
278 asking to enter a new tape each time.
280 4. Fixed a memory leak which was causing dump to grow very big
281 (270MB when dumping a 10GB filesystem...). Thanks to Jason
282 Fearon <jasonf@netrider.org.au> for reporting the bug.
284 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
285 ======================================================================
287 1. Make configure test if the system glob routines support
288 extended syntax (ALTDIRFUNC). If not, use the internal glob
289 routines instead of system ones. Thanks to Bernhard Sadlowski
290 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
291 and helping me resolve this and other minor libc5 compiling
294 2. Fix a problem when dumping a ext2fs with the 'filetype'
295 feature enabled. Thanks to Patrick J. LoPresti
296 <patl@cag.lcs.mit.edu> for reporting the bug and to
297 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
299 3. Made the nodump flag work on directories. A directory which
300 has the nodump flag gets never dumped, regardless of its
303 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
304 which allows dump on an active ext3 filesystem. However, this
305 is a "quick and dirty" patch which enables backup of an ext3
306 filesystem through the ext2 compatibility (by ignoring the
307 NEEDS_RECOVERY bit). The journal file is not recognized and
308 it is dumped (it should not).
310 5. Test the superblock compatibility flags when dumping, in order
311 to be sure that we know how to deal with specific features.
313 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
314 ====================================================================
316 1. Use lchown instead of chown, fixing a possible security problem
317 when restoring symlinks (a malicious user could use this
318 to deliberately corrupt the ownership of important system files).
319 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
320 this and providing the patch.
322 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
323 ====================================================================
325 1. Put dump sources under CVS, added Id tags in all files so
326 one can use 'ident' on binary files.
328 2. Added the dump/restore version in the usage text so one can
329 easily verify the version he is using.
331 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
332 a va_start/va_end problem on linux-ppc (always call va_start
333 va_end in pairs each time we use a vararg function).
335 4. Added again the DT_* constants because old libc does not
336 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
337 for submitting the bug report.
339 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
340 this should enable dumping big (huge) filesystems.
342 6. Added the RSH environment variable in order to be able to
343 use a rsh replacement like ssh when doing remote backups (and
344 bypass the security limitations of rcmd). Now you can do remote
345 backups without being root (or making dump setuid root).
347 7. Modified again the way dumpdates works. For incremental dumps,
348 we need to read dumpdates even if we are not using 'u' option.
349 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
352 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
353 ===================================================================
355 1. Removed the 'k' flag from the restore 'about' text if kerberos
358 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
359 to them (fsetflags takes a char*, setflags an open fd!).
361 3. (f)setflags is called only if the flags aren't empty. If the
362 file is a special file, a warning is printed, because changing
363 flags implies opening the device. Normally, a special file
364 should not have any flag... (Debian bug #29775, patch provided
365 by Abhijit Dasgupta <abhijit@ans.net>).
367 4. Made possible to dump a file system not mentioned in /etc/fstab.
368 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
370 5. Changed the default behaviour to not create dumpdates
371 unless 'u' option is specified. Removed the old "debian-patch"
372 which provided the same thing. (Debian bug #38136, #33818).
374 6. Removed all those dump*announce, since they were getting old...
376 7. Added warning messages if dumpdates does not exist and
377 when an inferior level dump does not exist (except for a level 0
380 8. Debugged the glob calls in interactive mode: restore used a
381 dirent struct which was different from the /usr/include/dirent.h
382 one (this used to work, is it a glibc2 change?), so none of the
383 compat glob (which used /usr/include/dirent.h) or the system glob
384 worked. Restore use now the system dirent (and the system
385 DT_* constants), which are compatible with BSD ones.
387 9. Added a configure flag (--with-dumpdatespath) to specify
388 the location of dumpdates file. By default, it is
391 10. Added the "AUTHOR" and "AVAILABILITY" sections and
392 included the current date/version in man pages.
394 11. Corrected the estimation of remaining time when
395 the operator doesn't change the tapes quickly enough. This
396 was an old bug, I thought I corrected it, and discovered
397 that in fact it was corrected in two different places, so
398 the results canceled each other...
400 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
401 ===================================================================
403 1. Integrated multiple patches from RedHat, Debian and SuSE:
405 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
407 - several fixes in the man pages.
408 - update the default tape device to /dev/st0.
409 - many updates for Linux Alpha (byte ordering, size_t etc).
411 - use environment variable for TMPDIR (instead of /tmp).
412 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
413 - workaround egcs bug (RedHat bugs #4281 and #2989).
414 - wire $(OPT) throughout Makefile's.
416 2. Upgrade the dump revision to 1, making possible to dump filesystems
417 made with e2fsprogs-1.15 or newer. Nothing seems to break...
419 3. Fix some compile warnings, prototype all functions.
421 4. Use glibc err/glob instead of internal compatibility
422 routines (only if available).
424 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
425 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
427 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
428 ======================================================================
430 1. Integrated the changes from FreeBSD-3.1-RELEASE
431 (mostly bug fixes, buffer overruns, dump has now an "automatic
432 tape length calculation" flag, dump/restore can use kerberos now
433 (this is NOT tested), use environment variables for TAPE and
436 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
438 3. Corrected a bug in restore when using 'C' option with multi-volumes
439 tapes (files splited accros two tapes give "size changed" errors
442 4. Corrected the long standing bug when dumping multiple tapes.
443 This works for me, needs further testing.
445 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
446 ====================================================================
448 1. Dump now runs correctly on kernels 2.1.x
449 Fix made by Gerald Peters <gapeters@worldnet.att.net>
451 Changes between versions 0.4b2 and 0.4b3
452 ========================================
454 1. Use realpath() if available
458 Changes between versions 0.4b1 and 0.4b2
459 ========================================
461 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
464 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
466 3. Fixed configure.in to correctly handle --enable-rmt
468 Changes between versions 0.3 and 0.4b1
469 ======================================
471 1. Integrated the changes from 4.4BSD-Lite2
473 2. Integrated the patches from Debian and Red Hat
475 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
477 4. Changed dump to use the Ext2fs library to get block addresses. This
478 should solve the endianness problem on SparcLinux.
480 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
481 distribution's one) to use autoconf
483 6. Fixed a few minor bugs
485 Changes between versions 0.2e and 0.2f
486 ======================================
488 1. Added the creation of named pipes (fifos) in restore.
490 2. Added the -N flag in the restore manual page.
492 3. Added the file kernel.patch which contains the llseek() optimization
493 patch for 1.2.x kernels.
495 4. Fixed a bug in the restoration of symbolic links: owner and group were
498 5. Integrated some changes from FreeBSD 2.2.
500 6. Added a call to ftruncate() after restoring each file to restore
501 correctly files ending by a hole.
503 Changes between versions 0.2d and 0.2e
504 ======================================
506 1. Fixed a bug in the "set owner/mode" process. Every file was restored
507 with owner = root (0) and group = root/wheel/whatever (0).
509 Changes between versions 0.2c and 0.2d
510 ======================================
512 1. Dump is now able to backup 2GB+ filesystems.
514 2. Dump and restore can now be linked as static binaries.
516 Changes between versions 0.2b and 0.2c
517 ======================================
519 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
521 Changes between versions 0.2a and 0.2b
522 ======================================
524 1. Really fixed the bug that I should have corrected in 0.2a.
526 2. Enabled optimization again.
528 Changes between versions 0.2 and 0.2a
529 =====================================
531 1. Disabled the optimization during compilation.
533 Changes between versions 0.1 and 0.2
534 ====================================
536 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
537 the fs_type field (actually, I modified the file fstab.c to make it
538 use the mntent functions).
540 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
541 made by dump should be readable by the BSD restore and backups made
542 by the BSD dump should be readable by restore. Unfortunately, this
543 means that the dump format has changed between version 0.1 and version
546 3. Dump is now able to backup a subtree, it is no longer limited to whole
547 filesystems like the BSD version.
549 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
552 Changes between versions 0.0 and 0.1
553 ====================================
555 1. Now create links rdump and rrestore during the `make install' step.
557 2. Linux port specific bugs added to the manual pages
559 3. Incorrect estimation of the number of tapes blocks fixed when doing
562 4. Better ls-like format in restore in interactive mode.