]> git.wh0rd.org - dump.git/blob - CHANGES
Error code to restore when compare errors detected.
[dump.git] / CHANGES
1 $Id: CHANGES,v 1.56 2000/05/28 16:51:13 stelian Exp $
2
3 Changes between versions 0.4b16 and 0.4b17 (released ??????????????)
4 ====================================================================
5
6 1. The -F script is called now *only* at the end of a tape,
7 not at the end of the dump. Thanks to Chris
8 <haul@informatik.tu-darmstadt.de> for the bug report.
9
10 2. Use posix signal handling to preserve dump functionality
11 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
12 the patch.
13
14 3. Made the exit code of restore in case of a 'C'ompare
15 command reflect the result of the compare. An exit status
16 of 0 means the dump archive is correct, 1 means tape errors,
17 2 means that some files were modified.
18
19 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
20 ====================================================================
21
22 1. Fixed some several duplicate 'DUMP: DUMP:' in the
23 output of dump.
24
25 2. Corrected the estimation of blocks to dump. Note that
26 this is still wrong for large directory inodes, due
27 to the size of a BSD directory entry on the tape
28 (estimation is based on the size of the ext2 directory,
29 which is smaller because it doesn't contain the null
30 character at the end).
31
32 3. Made dump report the total number of MB written to
33 tape at the end of run. Thanks to W. Reilly Cooley
34 <wcooley@nakedape.cc> for the patch.
35
36 4. Added the -X option to restore giving the possibility
37 to read the names of the files to be extracted/listed
38 from a text file (in addition of the command line).
39 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
40 patch.
41
42 5. Added the device name and the current volume number
43 as arguments to the end of tape script (dump -F option).
44
45 6. Made the multi-volume dump work again (after having
46 broken it in 0.4b15).
47
48 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
49 ===================================================================
50
51 1. Added a prompt command in interactive restore mode. Thanks
52 to Andreas Dilger <adilger@home.com> for the patch.
53
54 2. Fixed a buffer overflow problem in dump (caused by
55 not checking the size of the filesystem parameter).
56 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
57 reporting this on Bugtraq (and to several dump users
58 who forwarded me his mail).
59
60 3. Added the '-F script' option to dump in order to
61 launch a script at the end of each tape (to be used
62 with a tape changer, or to notify the sysadmin by
63 pager etc.).
64
65 4. Fixed a bug in restore compare code caused by the changes
66 I made in 0.4b14.
67
68 5. Fixed the treatment of options using the old BSD syntax
69 in both dump and restore.
70
71 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
72 =======================================================================
73
74 1. Fixed a bug in dump which may have caused invalid deleted
75 directories to be dumped out if they were deleted after the
76 mapping pass. This could occure on active filesystem and lead
77 to heap corruption (causing dump malfunction in many possible ways).
78 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
79 bug and submitting the patch.
80
81 2. Fixed the handling of the filesystem parameter in dump. A
82 '/mnt/disk' parameter caused the disk contents to be dumped,
83 but a '/mnt/disk/' parameter caused the mountpoint directory
84 to be dumped (generally an empty directory).
85
86 3. Improved the output of dump in order to tell which directory
87 it is currently dumping (when dumping a subtree).
88
89 4. Added the '-e' exclude inode option to dump. Thanks to
90 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
91
92 5. Added a REPORTING-BUGS file in order to provide a guide
93 on how to correctly report a bug in dump/restore.
94
95 6. Fixed a restore problem when restoring a hard link to an inode
96 having the immutable or append-only attribute set. Thanks to
97 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
98
99 7. Fixed a compatibility problem between dumps made on little
100 endian machines (the format was correct) and big endian
101 machines (the format was incorrect). This fix break the
102 compatibility with the older dumps made on big endian
103 machines (sparc, m86k, ppc etc). For the first time in
104 linux dump's history (I believe), the dumps made by *BSD,
105 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
106 so interchangeable. Thanks to Rob Cermak
107 <cermak@ahab.rutgers.edu> for submitting the bug and
108 helping me test the fix.
109
110 8. Fixed the way dump reports the remaining percent/time, if
111 the number of blocks actually dumped exceeds the estimated
112 number of blocks. Thanks to Jean-Paul van der Jagt
113 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
114
115 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
116 ======================================================================
117
118 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
119 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
120
121 2. Removed the requirement to build the RPM as root from the
122 spec file. Thanks to Christian Weisgerber
123 <naddy@mips.rhein-neckar.de> for submitting this.
124
125 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
126 causing dump to block when dumping really huge filesystems.
127 Many thanks to Patrik Schindler <poc@pocnet.net> for
128 helping me find this bug.
129
130 4. Fixed the treatment for an interrupt signal when dump access
131 the remote tape through RSH. Thanks to Christian Weisgerber
132 <naddy@mips.rhein-neckar.de> for providing the patch.
133
134 5. Fixed a bug which was causing dump/restore to display
135 garbage characters instead of the remote host name.
136
137 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
138 =====================================================================
139
140 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
141 <kukuk@suse.de> for submitting the patch.
142
143 2. Fix for the exit code when using the size estimate option of
144 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
145 submitting the patch.
146
147 3. Handle EINTR in atomical reads/writes in dump, which was causing
148 dump to fail on some systems. Thanks to Eric Jergensen
149 <eric@dvns.com> for reporting the bug and submitting the patch.
150
151 4. Handle more than 16 characters for the device names in dumpdates.
152 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
153 tracking down the problem and proposing the solution.
154
155 5. Fixed a bug in dump which prevented the creation of the
156 dumpdates file when doing a 0-level dump without already
157 having a dumpdates file. Thanks to Patrik Schindler
158 <poc@pocnet.net> for reporting the bug.
159
160 6. Changed the way dump 'S' flag reports the size estimate
161 from number of blocks into bytes (making it compatible
162 with the Solaris version, and simplifying things for
163 amanda users). Thanks to Jason L Tibbitts III
164 <tibbs@math.uh.edu> for reporting the bug.
165
166 7. Fixed a compatibility problem in linux/alpha dump tape format.
167 Now the linux/alpha dump are (again) compatible with the
168 other dump formats. But this breaks compatibility with
169 older dumps made on alpha. Thanks to Mike Tibor
170 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
171
172 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
173 ======================================================================
174
175 1. Added a '--enable-kerberos' to configure.
176
177 2. Added a 'S' option to dump which determines the amount of space
178 that is needed to perform the dump without actually doing it, similar
179 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
180 <cermak@ahab.rutgers.edu>.
181
182 3. Added a 'M' multi-volume option to dump and restore which enables
183 dumping to multiple files (useful when dumping to an ext2
184 partition to make several dump files in order to bypass the 2GB
185 file size limitation). The 'f' argument is treated as a prefix and
186 the output files will be named <prefix>001, <prefix>002 etc. With
187 the 'M' flag, restore automatically selects the right file without
188 asking to enter a new tape each time.
189
190 4. Fixed a memory leak which was causing dump to grow very big
191 (270MB when dumping a 10GB filesystem...). Thanks to Jason
192 Fearon <jasonf@netrider.org.au> for reporting the bug.
193
194 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
195 ======================================================================
196
197 1. Make configure test if the system glob routines support
198 extended syntax (ALTDIRFUNC). If not, use the internal glob
199 routines instead of system ones. Thanks to Bernhard Sadlowski
200 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
201 and helping me resolve this and other minor libc5 compiling
202 glitches.
203
204 2. Fix a problem when dumping a ext2fs with the 'filetype'
205 feature enabled. Thanks to Patrick J. LoPresti
206 <patl@cag.lcs.mit.edu> for reporting the bug and to
207 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
208
209 3. Made the nodump flag work on directories. A directory which
210 has the nodump flag gets never dumped, regardless of its
211 contents.
212
213 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
214 which allows dump on an active ext3 filesystem. However, this
215 is a "quick and dirty" patch which enables backup of an ext3
216 filesystem through the ext2 compatibility (by ignoring the
217 NEEDS_RECOVERY bit). The journal file is not recognized and
218 it is dumped (it should not).
219
220 5. Test the superblock compatibility flags when dumping, in order
221 to be sure that we know how to deal with specific features.
222
223 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
224 ====================================================================
225
226 1. Use lchown instead of chown, fixing a possible security problem
227 when restoring symlinks (a malicious user could use this
228 to deliberately corrupt the ownership of important system files).
229 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
230 this and providing the patch.
231
232 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
233 ====================================================================
234
235 1. Put dump sources under CVS, added Id tags in all files so
236 one can use 'ident' on binary files.
237
238 2. Added the dump/restore version in the usage text so one can
239 easily verify the version he is using.
240
241 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
242 a va_start/va_end problem on linux-ppc (always call va_start
243 va_end in pairs each time we use a vararg function).
244
245 4. Added again the DT_* constants because old libc does not
246 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
247 for submitting the bug report.
248
249 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
250 this should enable dumping big (huge) filesystems.
251
252 6. Added the RSH environment variable in order to be able to
253 use a rsh replacement like ssh when doing remote backups (and
254 bypass the security limitations of rcmd). Now you can do remote
255 backups without being root (or making dump setuid root).
256
257 7. Modified again the way dumpdates works. For incremental dumps,
258 we need to read dumpdates even if we are not using 'u' option.
259 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
260 this should work.
261
262 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
263 ===================================================================
264
265 1. Removed the 'k' flag from the restore 'about' text if kerberos
266 was not compiled in.
267
268 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
269 to them (fsetflags takes a char*, setflags an open fd!).
270
271 3. (f)setflags is called only if the flags aren't empty. If the
272 file is a special file, a warning is printed, because changing
273 flags implies opening the device. Normally, a special file
274 should not have any flag... (Debian bug #29775, patch provided
275 by Abhijit Dasgupta <abhijit@ans.net>).
276
277 4. Made possible to dump a file system not mentioned in /etc/fstab.
278 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
279
280 5. Changed the default behaviour to not create dumpdates
281 unless 'u' option is specified. Removed the old "debian-patch"
282 which provided the same thing. (Debian bug #38136, #33818).
283
284 6. Removed all those dump*announce, since they were getting old...
285
286 7. Added warning messages if dumpdates does not exist and
287 when an inferior level dump does not exist (except for a level 0
288 dump).
289
290 8. Debugged the glob calls in interactive mode: restore used a
291 dirent struct which was different from the /usr/include/dirent.h
292 one (this used to work, is it a glibc2 change?), so none of the
293 compat glob (which used /usr/include/dirent.h) or the system glob
294 worked. Restore use now the system dirent (and the system
295 DT_* constants), which are compatible with BSD ones.
296
297 9. Added a configure flag (--with-dumpdatespath) to specify
298 the location of dumpdates file. By default, it is
299 /etc/dumpdates.
300
301 10. Added the "AUTHOR" and "AVAILABILITY" sections and
302 included the current date/version in man pages.
303
304 11. Corrected the estimation of remaining time when
305 the operator doesn't change the tapes quickly enough. This
306 was an old bug, I thought I corrected it, and discovered
307 that in fact it was corrected in two different places, so
308 the results canceled each other...
309
310 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
311 ===================================================================
312
313 1. Integrated multiple patches from RedHat, Debian and SuSE:
314
315 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
316 isn't specified.
317 - several fixes in the man pages.
318 - update the default tape device to /dev/st0.
319 - many updates for Linux Alpha (byte ordering, size_t etc).
320 - buffer overruns.
321 - use environment variable for TMPDIR (instead of /tmp).
322 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
323 - workaround egcs bug (RedHat bugs #4281 and #2989).
324 - wire $(OPT) throughout Makefile's.
325
326 2. Upgrade the dump revision to 1, making possible to dump filesystems
327 made with e2fsprogs-1.15 or newer. Nothing seems to break...
328
329 3. Fix some compile warnings, prototype all functions.
330
331 4. Use glibc err/glob instead of internal compatibility
332 routines (only if available).
333
334 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
335 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
336
337 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
338 ======================================================================
339
340 1. Integrated the changes from FreeBSD-3.1-RELEASE
341 (mostly bug fixes, buffer overruns, dump has now an "automatic
342 tape length calculation" flag, dump/restore can use kerberos now
343 (this is NOT tested), use environment variables for TAPE and
344 RMT etc.).
345
346 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
347
348 3. Corrected a bug in restore when using 'C' option with multi-volumes
349 tapes (files splited accros two tapes give "size changed" errors
350 when comparing).
351
352 4. Corrected the long standing bug when dumping multiple tapes.
353 This works for me, needs further testing.
354
355 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
356 ====================================================================
357
358 1. Dump now runs correctly on kernels 2.1.x
359 Fix made by Gerald Peters <gapeters@worldnet.att.net>
360
361 Changes between versions 0.4b2 and 0.4b3
362 ========================================
363
364 1. Use realpath() if available
365
366 2. Report statistics
367
368 Changes between versions 0.4b1 and 0.4b2
369 ========================================
370
371 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
372 the patch)
373
374 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
375
376 3. Fixed configure.in to correctly handle --enable-rmt
377
378 Changes between versions 0.3 and 0.4b1
379 ======================================
380
381 1. Integrated the changes from 4.4BSD-Lite2
382
383 2. Integrated the patches from Debian and Red Hat
384
385 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
386
387 4. Changed dump to use the Ext2fs library to get block addresses. This
388 should solve the endianness problem on SparcLinux.
389
390 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
391 distribution's one) to use autoconf
392
393 6. Fixed a few minor bugs
394
395 Changes between versions 0.2e and 0.2f
396 ======================================
397
398 1. Added the creation of named pipes (fifos) in restore.
399
400 2. Added the -N flag in the restore manual page.
401
402 3. Added the file kernel.patch which contains the llseek() optimization
403 patch for 1.2.x kernels.
404
405 4. Fixed a bug in the restoration of symbolic links: owner and group were
406 not restored.
407
408 5. Integrated some changes from FreeBSD 2.2.
409
410 6. Added a call to ftruncate() after restoring each file to restore
411 correctly files ending by a hole.
412
413 Changes between versions 0.2d and 0.2e
414 ======================================
415
416 1. Fixed a bug in the "set owner/mode" process. Every file was restored
417 with owner = root (0) and group = root/wheel/whatever (0).
418
419 Changes between versions 0.2c and 0.2d
420 ======================================
421
422 1. Dump is now able to backup 2GB+ filesystems.
423
424 2. Dump and restore can now be linked as static binaries.
425
426 Changes between versions 0.2b and 0.2c
427 ======================================
428
429 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
430
431 Changes between versions 0.2a and 0.2b
432 ======================================
433
434 1. Really fixed the bug that I should have corrected in 0.2a.
435
436 2. Enabled optimization again.
437
438 Changes between versions 0.2 and 0.2a
439 =====================================
440
441 1. Disabled the optimization during compilation.
442
443 Changes between versions 0.1 and 0.2
444 ====================================
445
446 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
447 the fs_type field (actually, I modified the file fstab.c to make it
448 use the mntent functions).
449
450 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
451 made by dump should be readable by the BSD restore and backups made
452 by the BSD dump should be readable by restore. Unfortunately, this
453 means that the dump format has changed between version 0.1 and version
454 0.2 :-(
455
456 3. Dump is now able to backup a subtree, it is no longer limited to whole
457 filesystems like the BSD version.
458
459 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
460 than 2 GB.
461
462 Changes between versions 0.0 and 0.1
463 ====================================
464
465 1. Now create links rdump and rrestore during the `make install' step.
466
467 2. Linux port specific bugs added to the manual pages
468
469 3. Incorrect estimation of the number of tapes blocks fixed when doing
470 incremental backups.
471
472 4. Better ls-like format in restore in interactive mode.