]> git.wh0rd.org - dump.git/blob - CHANGES
Update to the latest RedHat RPM spec file.
[dump.git] / CHANGES
1 $Id: CHANGES,v 1.75 2000/08/20 15:46:52 stelian Exp $
2
3 Changes between versions 0.4b17 and 0.4b19 (released ?????????????)
4 ===================================================================
5
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).
9
10 2. Specified the default blocksize in the dump man page.
11
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.
16
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.
21
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.
26
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
31 reporting the bug.
32
33 7. Update the RPM spec file to the latest RedHat version,
34 providing FHS packaging and other cosmetic changes.
35
36 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
37 ===================================================================
38
39 1. Fixed a potential buffer overflow in restore. Thanks
40 to Stan Bubrouski <satan@fastdial.net> for reporting
41 the bug.
42
43 2. Fixed a readline-related bug which prevented
44 'cat DUMPFILE | restore -i -f -' from working. Thanks
45 to Charles Karney <karney@users.sourceforge.net>
46 for the bug report.
47
48 3. Changed a few "panic" into "exit", causing restore to
49 be more stable against some attacks (like the last one
50 reported on Bugtraq, although the last version of restore
51 was not vulnerable - just dumped core). Thanks to
52 Andreas Hasenack <andreas@conectiva.com.br> for reporting
53 the bugs to me.
54
55 4. Removed the suid-root bit on dump and restore in the
56 default build (and generated RPMs). It should be safer
57 now. Users who need the suid-root capabilities in order
58 to make network backups should read first the man page
59 of dump and enable the suid bit manually.
60
61 5. Added -ltermcap to the compile parameters for restore
62 when using readline support, in order to make the compile
63 process work on some systems (SuSE ?). Thanks to
64 Patrik Schindler <poc@pocnet.net> for reporting the bug.
65
66 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
67 ====================================================================
68
69 1. The -F script is called now *only* at the end of a tape,
70 not at the end of the dump. Thanks to Christian Haul
71 <haul@informatik.tu-darmstadt.de> for the bug report.
72
73 Normally, the device name and the current volume number
74 are passed on the command line of the script. However,
75 if you want the old style script (with no arguments
76 passed on the command line), you can enable it in
77 configure with the --enable-oldstylefscript.
78
79 2. Use posix signal handling to preserve dump functionality
80 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
81 the patch.
82
83 3. Made the exit code of restore in case of a 'C'ompare
84 command reflect the result of the compare. An exit status
85 of 0 means the dump archive is correct, 1 means tape errors,
86 2 means that some files were modified. Thanks to Kenneth Porter
87 <shiva@well.com> for the suggestion.
88
89 4. Made (finally) quotes work in interactive restore.
90
91 5. Small fixes in order to allow dump to compile with a
92 really old e2fsprogs version. Thanks to Ian Zimmerman
93 <itz@speakeasy.org> for the bug report.
94
95 6. Add GNU readline capabilities to interactive restore.
96 Use configure's --enable-readline flag to enable this feature.
97 Thanks to Patrik Schindler <poc@pocnet.net> for the
98 suggestion.
99
100 7. Do the compare on the fly in restore 'C' mode (this will
101 allow not to exhaust the available /tmp space when
102 ccmparing large files). Thanks to Kenneth Porter
103 <shiva@well.com> for the suggestion.
104
105 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
106 ====================================================================
107
108 1. Fixed some several duplicate 'DUMP: DUMP:' in the
109 output of dump.
110
111 2. Corrected the estimation of blocks to dump. Note that
112 this is still wrong for large directory inodes, due
113 to the size of a BSD directory entry on the tape
114 (estimation is based on the size of the ext2 directory,
115 which is smaller because it doesn't contain the null
116 character at the end).
117
118 3. Made dump report the total number of MB written to
119 tape at the end of run. Thanks to W. Reilly Cooley
120 <wcooley@nakedape.cc> for the patch.
121
122 4. Added the -X option to restore giving the possibility
123 to read the names of the files to be extracted/listed
124 from a text file (in addition of the command line).
125 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
126 patch.
127
128 5. Added the device name and the current volume number
129 as arguments to the end of tape script (dump -F option).
130
131 6. Made the multi-volume dump work again (after having
132 broken it in 0.4b15).
133
134 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
135 ===================================================================
136
137 1. Added a prompt command in interactive restore mode. Thanks
138 to Andreas Dilger <adilger@home.com> for the patch.
139
140 2. Fixed a buffer overflow problem in dump (caused by
141 not checking the size of the filesystem parameter).
142 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
143 reporting this on Bugtraq (and to several dump users
144 who forwarded me his mail).
145
146 3. Added the '-F script' option to dump in order to
147 launch a script at the end of each tape (to be used
148 with a tape changer, or to notify the sysadmin by
149 pager etc.).
150
151 4. Fixed a bug in restore compare code caused by the changes
152 I made in 0.4b14.
153
154 5. Fixed the treatment of options using the old BSD syntax
155 in both dump and restore.
156
157 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
158 =======================================================================
159
160 1. Fixed a bug in dump which may have caused invalid deleted
161 directories to be dumped out if they were deleted after the
162 mapping pass. This could occure on active filesystem and lead
163 to heap corruption (causing dump malfunction in many possible ways).
164 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
165 bug and submitting the patch.
166
167 2. Fixed the handling of the filesystem parameter in dump. A
168 '/mnt/disk' parameter caused the disk contents to be dumped,
169 but a '/mnt/disk/' parameter caused the mountpoint directory
170 to be dumped (generally an empty directory).
171
172 3. Improved the output of dump in order to tell which directory
173 it is currently dumping (when dumping a subtree).
174
175 4. Added the '-e' exclude inode option to dump. Thanks to
176 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
177
178 5. Added a REPORTING-BUGS file in order to provide a guide
179 on how to correctly report a bug in dump/restore.
180
181 6. Fixed a restore problem when restoring a hard link to an inode
182 having the immutable or append-only attribute set. Thanks to
183 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
184
185 7. Fixed a compatibility problem between dumps made on little
186 endian machines (the format was correct) and big endian
187 machines (the format was incorrect). This fix break the
188 compatibility with the older dumps made on big endian
189 machines (sparc, m86k, ppc etc). For the first time in
190 linux dump's history (I believe), the dumps made by *BSD,
191 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
192 so interchangeable. Thanks to Rob Cermak
193 <cermak@ahab.rutgers.edu> for submitting the bug and
194 helping me test the fix.
195
196 8. Fixed the way dump reports the remaining percent/time, if
197 the number of blocks actually dumped exceeds the estimated
198 number of blocks. Thanks to Jean-Paul van der Jagt
199 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
200
201 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
202 ======================================================================
203
204 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
205 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
206
207 2. Removed the requirement to build the RPM as root from the
208 spec file. Thanks to Christian Weisgerber
209 <naddy@mips.rhein-neckar.de> for submitting this.
210
211 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
212 causing dump to block when dumping really huge filesystems.
213 Many thanks to Patrik Schindler <poc@pocnet.net> for
214 helping me find this bug.
215
216 4. Fixed the treatment for an interrupt signal when dump access
217 the remote tape through RSH. Thanks to Christian Weisgerber
218 <naddy@mips.rhein-neckar.de> for providing the patch.
219
220 5. Fixed a bug which was causing dump/restore to display
221 garbage characters instead of the remote host name.
222
223 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
224 =====================================================================
225
226 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
227 <kukuk@suse.de> for submitting the patch.
228
229 2. Fix for the exit code when using the size estimate option of
230 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
231 submitting the patch.
232
233 3. Handle EINTR in atomical reads/writes in dump, which was causing
234 dump to fail on some systems. Thanks to Eric Jergensen
235 <eric@dvns.com> for reporting the bug and submitting the patch.
236
237 4. Handle more than 16 characters for the device names in dumpdates.
238 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
239 tracking down the problem and proposing the solution.
240
241 5. Fixed a bug in dump which prevented the creation of the
242 dumpdates file when doing a 0-level dump without already
243 having a dumpdates file. Thanks to Patrik Schindler
244 <poc@pocnet.net> for reporting the bug.
245
246 6. Changed the way dump 'S' flag reports the size estimate
247 from number of blocks into bytes (making it compatible
248 with the Solaris version, and simplifying things for
249 amanda users). Thanks to Jason L Tibbitts III
250 <tibbs@math.uh.edu> for reporting the bug.
251
252 7. Fixed a compatibility problem in linux/alpha dump tape format.
253 Now the linux/alpha dump are (again) compatible with the
254 other dump formats. But this breaks compatibility with
255 older dumps made on alpha. Thanks to Mike Tibor
256 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
257
258 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
259 ======================================================================
260
261 1. Added a '--enable-kerberos' to configure.
262
263 2. Added a 'S' option to dump which determines the amount of space
264 that is needed to perform the dump without actually doing it, similar
265 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
266 <cermak@ahab.rutgers.edu>.
267
268 3. Added a 'M' multi-volume option to dump and restore which enables
269 dumping to multiple files (useful when dumping to an ext2
270 partition to make several dump files in order to bypass the 2GB
271 file size limitation). The 'f' argument is treated as a prefix and
272 the output files will be named <prefix>001, <prefix>002 etc. With
273 the 'M' flag, restore automatically selects the right file without
274 asking to enter a new tape each time.
275
276 4. Fixed a memory leak which was causing dump to grow very big
277 (270MB when dumping a 10GB filesystem...). Thanks to Jason
278 Fearon <jasonf@netrider.org.au> for reporting the bug.
279
280 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
281 ======================================================================
282
283 1. Make configure test if the system glob routines support
284 extended syntax (ALTDIRFUNC). If not, use the internal glob
285 routines instead of system ones. Thanks to Bernhard Sadlowski
286 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
287 and helping me resolve this and other minor libc5 compiling
288 glitches.
289
290 2. Fix a problem when dumping a ext2fs with the 'filetype'
291 feature enabled. Thanks to Patrick J. LoPresti
292 <patl@cag.lcs.mit.edu> for reporting the bug and to
293 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
294
295 3. Made the nodump flag work on directories. A directory which
296 has the nodump flag gets never dumped, regardless of its
297 contents.
298
299 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
300 which allows dump on an active ext3 filesystem. However, this
301 is a "quick and dirty" patch which enables backup of an ext3
302 filesystem through the ext2 compatibility (by ignoring the
303 NEEDS_RECOVERY bit). The journal file is not recognized and
304 it is dumped (it should not).
305
306 5. Test the superblock compatibility flags when dumping, in order
307 to be sure that we know how to deal with specific features.
308
309 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
310 ====================================================================
311
312 1. Use lchown instead of chown, fixing a possible security problem
313 when restoring symlinks (a malicious user could use this
314 to deliberately corrupt the ownership of important system files).
315 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
316 this and providing the patch.
317
318 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
319 ====================================================================
320
321 1. Put dump sources under CVS, added Id tags in all files so
322 one can use 'ident' on binary files.
323
324 2. Added the dump/restore version in the usage text so one can
325 easily verify the version he is using.
326
327 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
328 a va_start/va_end problem on linux-ppc (always call va_start
329 va_end in pairs each time we use a vararg function).
330
331 4. Added again the DT_* constants because old libc does not
332 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
333 for submitting the bug report.
334
335 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
336 this should enable dumping big (huge) filesystems.
337
338 6. Added the RSH environment variable in order to be able to
339 use a rsh replacement like ssh when doing remote backups (and
340 bypass the security limitations of rcmd). Now you can do remote
341 backups without being root (or making dump setuid root).
342
343 7. Modified again the way dumpdates works. For incremental dumps,
344 we need to read dumpdates even if we are not using 'u' option.
345 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
346 this should work.
347
348 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
349 ===================================================================
350
351 1. Removed the 'k' flag from the restore 'about' text if kerberos
352 was not compiled in.
353
354 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
355 to them (fsetflags takes a char*, setflags an open fd!).
356
357 3. (f)setflags is called only if the flags aren't empty. If the
358 file is a special file, a warning is printed, because changing
359 flags implies opening the device. Normally, a special file
360 should not have any flag... (Debian bug #29775, patch provided
361 by Abhijit Dasgupta <abhijit@ans.net>).
362
363 4. Made possible to dump a file system not mentioned in /etc/fstab.
364 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
365
366 5. Changed the default behaviour to not create dumpdates
367 unless 'u' option is specified. Removed the old "debian-patch"
368 which provided the same thing. (Debian bug #38136, #33818).
369
370 6. Removed all those dump*announce, since they were getting old...
371
372 7. Added warning messages if dumpdates does not exist and
373 when an inferior level dump does not exist (except for a level 0
374 dump).
375
376 8. Debugged the glob calls in interactive mode: restore used a
377 dirent struct which was different from the /usr/include/dirent.h
378 one (this used to work, is it a glibc2 change?), so none of the
379 compat glob (which used /usr/include/dirent.h) or the system glob
380 worked. Restore use now the system dirent (and the system
381 DT_* constants), which are compatible with BSD ones.
382
383 9. Added a configure flag (--with-dumpdatespath) to specify
384 the location of dumpdates file. By default, it is
385 /etc/dumpdates.
386
387 10. Added the "AUTHOR" and "AVAILABILITY" sections and
388 included the current date/version in man pages.
389
390 11. Corrected the estimation of remaining time when
391 the operator doesn't change the tapes quickly enough. This
392 was an old bug, I thought I corrected it, and discovered
393 that in fact it was corrected in two different places, so
394 the results canceled each other...
395
396 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
397 ===================================================================
398
399 1. Integrated multiple patches from RedHat, Debian and SuSE:
400
401 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
402 isn't specified.
403 - several fixes in the man pages.
404 - update the default tape device to /dev/st0.
405 - many updates for Linux Alpha (byte ordering, size_t etc).
406 - buffer overruns.
407 - use environment variable for TMPDIR (instead of /tmp).
408 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
409 - workaround egcs bug (RedHat bugs #4281 and #2989).
410 - wire $(OPT) throughout Makefile's.
411
412 2. Upgrade the dump revision to 1, making possible to dump filesystems
413 made with e2fsprogs-1.15 or newer. Nothing seems to break...
414
415 3. Fix some compile warnings, prototype all functions.
416
417 4. Use glibc err/glob instead of internal compatibility
418 routines (only if available).
419
420 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
421 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
422
423 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
424 ======================================================================
425
426 1. Integrated the changes from FreeBSD-3.1-RELEASE
427 (mostly bug fixes, buffer overruns, dump has now an "automatic
428 tape length calculation" flag, dump/restore can use kerberos now
429 (this is NOT tested), use environment variables for TAPE and
430 RMT etc.).
431
432 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
433
434 3. Corrected a bug in restore when using 'C' option with multi-volumes
435 tapes (files splited accros two tapes give "size changed" errors
436 when comparing).
437
438 4. Corrected the long standing bug when dumping multiple tapes.
439 This works for me, needs further testing.
440
441 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
442 ====================================================================
443
444 1. Dump now runs correctly on kernels 2.1.x
445 Fix made by Gerald Peters <gapeters@worldnet.att.net>
446
447 Changes between versions 0.4b2 and 0.4b3
448 ========================================
449
450 1. Use realpath() if available
451
452 2. Report statistics
453
454 Changes between versions 0.4b1 and 0.4b2
455 ========================================
456
457 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
458 the patch)
459
460 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
461
462 3. Fixed configure.in to correctly handle --enable-rmt
463
464 Changes between versions 0.3 and 0.4b1
465 ======================================
466
467 1. Integrated the changes from 4.4BSD-Lite2
468
469 2. Integrated the patches from Debian and Red Hat
470
471 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
472
473 4. Changed dump to use the Ext2fs library to get block addresses. This
474 should solve the endianness problem on SparcLinux.
475
476 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
477 distribution's one) to use autoconf
478
479 6. Fixed a few minor bugs
480
481 Changes between versions 0.2e and 0.2f
482 ======================================
483
484 1. Added the creation of named pipes (fifos) in restore.
485
486 2. Added the -N flag in the restore manual page.
487
488 3. Added the file kernel.patch which contains the llseek() optimization
489 patch for 1.2.x kernels.
490
491 4. Fixed a bug in the restoration of symbolic links: owner and group were
492 not restored.
493
494 5. Integrated some changes from FreeBSD 2.2.
495
496 6. Added a call to ftruncate() after restoring each file to restore
497 correctly files ending by a hole.
498
499 Changes between versions 0.2d and 0.2e
500 ======================================
501
502 1. Fixed a bug in the "set owner/mode" process. Every file was restored
503 with owner = root (0) and group = root/wheel/whatever (0).
504
505 Changes between versions 0.2c and 0.2d
506 ======================================
507
508 1. Dump is now able to backup 2GB+ filesystems.
509
510 2. Dump and restore can now be linked as static binaries.
511
512 Changes between versions 0.2b and 0.2c
513 ======================================
514
515 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
516
517 Changes between versions 0.2a and 0.2b
518 ======================================
519
520 1. Really fixed the bug that I should have corrected in 0.2a.
521
522 2. Enabled optimization again.
523
524 Changes between versions 0.2 and 0.2a
525 =====================================
526
527 1. Disabled the optimization during compilation.
528
529 Changes between versions 0.1 and 0.2
530 ====================================
531
532 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
533 the fs_type field (actually, I modified the file fstab.c to make it
534 use the mntent functions).
535
536 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
537 made by dump should be readable by the BSD restore and backups made
538 by the BSD dump should be readable by restore. Unfortunately, this
539 means that the dump format has changed between version 0.1 and version
540 0.2 :-(
541
542 3. Dump is now able to backup a subtree, it is no longer limited to whole
543 filesystems like the BSD version.
544
545 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
546 than 2 GB.
547
548 Changes between versions 0.0 and 0.1
549 ====================================
550
551 1. Now create links rdump and rrestore during the `make install' step.
552
553 2. Linux port specific bugs added to the manual pages
554
555 3. Incorrect estimation of the number of tapes blocks fixed when doing
556 incremental backups.
557
558 4. Better ls-like format in restore in interactive mode.