]> git.wh0rd.org - dump.git/blob - CHANGES
Added REPORTING-BUGS.
[dump.git] / CHANGES
1 $Id: CHANGES,v 1.40 2000/02/04 20:57:31 stelian Exp $
2
3 Changes between versions 0.4b13 and 0.4b14 (released ????????????????)
4 ======================================================================
5
6 1. Fixed a bug in dump which may have caused invalid deleted
7 directories to be dumped out if they were deleted after the
8 mapping pass. This could occure on active filesystem and lead
9 to heap corruption (causing dump malfunction in many possible ways).
10 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
11 bug and submitting the patch.
12
13 2. Fixed the handling of the filesystem parameter in dump. A
14 '/mnt/disk' parameter caused the disk contents to be dumped,
15 but a '/mnt/disk/' parameter caused the mountpoint directory
16 to be dumped (generally an empty directory).
17
18 3. Improved the output of dump in order to tell which directory
19 it is currently dumping (when dumping a subtree).
20
21 4. Added the '-e' exclude inode option to dump. Thanks to
22 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
23
24 5. Added a REPORTING-BUGS file in order to provide a guide
25 on how to correctly report a bug in dump/restore.
26
27 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
28 ======================================================================
29
30 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
31 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
32
33 2. Removed the requirement to build the RPM as root from the
34 spec file. Thanks to Christian Weisgerber
35 <naddy@mips.rhein-neckar.de> for submitting this.
36
37 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
38 causing dump to block when dumping really huge filesystems.
39 Many thanks to Patrik Schindler <poc@pocnet.net> for
40 helping me find this bug.
41
42 4. Fixed the treatment for an interrupt signal when dump access
43 the remote tape through RSH. Thanks to Christian Weisgerber
44 <naddy@mips.rhein-neckar.de> for providing the patch.
45
46 5. Fixed a bug which was causing dump/restore to display
47 garbage characters instead of the remote host name.
48
49 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
50 =====================================================================
51
52 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
53 <kukuk@suse.de> for submitting the patch.
54
55 2. Fix for the exit code when using the size estimate option of
56 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
57 submitting the patch.
58
59 3. Handle EINTR in atomical reads/writes in dump, which was causing
60 dump to fail on some systems. Thanks to Eric Jergensen
61 <eric@dvns.com> for reporting the bug and submitting the patch.
62
63 4. Handle more than 16 characters for the device names in dumpdates.
64 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
65 tracking down the problem and proposing the solution.
66
67 5. Fixed a bug in dump which prevented the creation of the
68 dumpdates file when doing a 0-level dump without already
69 having a dumpdates file. Thanks to Patrik Schindler
70 <poc@pocnet.net> for reporting the bug.
71
72 6. Changed the way dump 'S' flag reports the size estimate
73 from number of blocks into bytes (making it compatible
74 with the Solaris version, and simplifying things for
75 amanda users). Thanks to Jason L Tibbitts III
76 <tibbs@math.uh.edu> for reporting the bug.
77
78 7. Fixed a compatibility problem in linux/alpha dump tape format.
79 Now the linux/alpha dump are (again) compatible with the
80 other dump formats. But this breaks compatibility with
81 older dumps made on alpha. Thanks to Mike Tibor
82 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
83
84 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
85 ======================================================================
86
87 1. Added a '--enable-kerberos' to configure.
88
89 2. Added a 'S' option to dump which determines the amount of space
90 that is needed to perform the dump without actually doing it, similar
91 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
92 <cermak@ahab.rutgers.edu>.
93
94 3. Added a 'M' multi-volume option to dump and restore which enables
95 dumping to multiple files (useful when dumping to an ext2
96 partition to make several dump files in order to bypass the 2GB
97 file size limitation). The 'f' argument is treated as a prefix and
98 the output files will be named <prefix>001, <prefix>002 etc. With
99 the 'M' flag, restore automatically selects the right file without
100 asking to enter a new tape each time.
101
102 4. Fixed a memory leak which was causing dump to grow very big
103 (270MB when dumping a 10GB filesystem...). Thanks to Jason
104 Fearon <jasonf@netrider.org.au> for reporting the bug.
105
106 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
107 ======================================================================
108
109 1. Make configure test if the system glob routines support
110 extended syntax (ALTDIRFUNC). If not, use the internal glob
111 routines instead of system ones. Thanks to Bernhard Sadlowski
112 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
113 and helping me resolve this and other minor libc5 compiling
114 glitches.
115
116 2. Fix a problem when dumping a ext2fs with the 'filetype'
117 feature enabled. Thanks to Patrick J. LoPresti
118 <patl@cag.lcs.mit.edu> for reporting the bug and to
119 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
120
121 3. Made the nodump flag work on directories. A directory which
122 has the nodump flag gets never dumped, regardless of its
123 contents.
124
125 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
126 which allows dump on an active ext3 filesystem. However, this
127 is a "quick and dirty" patch which enables backup of an ext3
128 filesystem through the ext2 compatibility (by ignoring the
129 NEEDS_RECOVERY bit). The journal file is not recognized and
130 it is dumped (it should not).
131
132 5. Test the superblock compatibility flags when dumping, in order
133 to be sure that we know how to deal with specific features.
134
135 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
136 ====================================================================
137
138 1. Use lchown instead of chown, fixing a possible security problem
139 when restoring symlinks (a malicious user could use this
140 to deliberately corrupt the ownership of important system files).
141 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
142 this and providing the patch.
143
144 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
145 ====================================================================
146
147 1. Put dump sources under CVS, added Id tags in all files so
148 one can use 'ident' on binary files.
149
150 2. Added the dump/restore version in the usage text so one can
151 easily verify the version he is using.
152
153 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
154 a va_start/va_end problem on linux-ppc (always call va_start
155 va_end in pairs each time we use a vararg function).
156
157 4. Added again the DT_* constants because old libc does not
158 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
159 for submitting the bug report.
160
161 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
162 this should enable dumping big (huge) filesystems.
163
164 6. Added the RSH environment variable in order to be able to
165 use a rsh replacement like ssh when doing remote backups (and
166 bypass the security limitations of rcmd). Now you can do remote
167 backups without being root (or making dump setuid root).
168
169 7. Modified again the way dumpdates works. For incremental dumps,
170 we need to read dumpdates even if we are not using 'u' option.
171 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
172 this should work.
173
174 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
175 ===================================================================
176
177 1. Removed the 'k' flag from the restore 'about' text if kerberos
178 was not compiled in.
179
180 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
181 to them (fsetflags takes a char*, setflags an open fd!).
182
183 3. (f)setflags is called only if the flags aren't empty. If the
184 file is a special file, a warning is printed, because changing
185 flags implies opening the device. Normally, a special file
186 should not have any flag... (Debian bug #29775, patch provided
187 by Abhijit Dasgupta <abhijit@ans.net>).
188
189 4. Made possible to dump a file system not mentioned in /etc/fstab.
190 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
191
192 5. Changed the default behaviour to not create dumpdates
193 unless 'u' option is specified. Removed the old "debian-patch"
194 which provided the same thing. (Debian bug #38136, #33818).
195
196 6. Removed all those dump*announce, since they were getting old...
197
198 7. Added warning messages if dumpdates does not exist and
199 when an inferior level dump does not exist (except for a level 0
200 dump).
201
202 8. Debugged the glob calls in interactive mode: restore used a
203 dirent struct which was different from the /usr/include/dirent.h
204 one (this used to work, is it a glibc2 change?), so none of the
205 compat glob (which used /usr/include/dirent.h) or the system glob
206 worked. Restore use now the system dirent (and the system
207 DT_* constants), which are compatible with BSD ones.
208
209 9. Added a configure flag (--with-dumpdatespath) to specify
210 the location of dumpdates file. By default, it is
211 /etc/dumpdates.
212
213 10. Added the "AUTHOR" and "AVAILABILITY" sections and
214 included the current date/version in man pages.
215
216 11. Corrected the estimation of remaining time when
217 the operator doesn't change the tapes quickly enough. This
218 was an old bug, I thought I corrected it, and discovered
219 that in fact it was corrected in two different places, so
220 the results canceled each other...
221
222 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
223 ===================================================================
224
225 1. Integrated multiple patches from RedHat, Debian and SuSE:
226
227 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
228 isn't specified.
229 - several fixes in the man pages.
230 - update the default tape device to /dev/st0.
231 - many updates for Linux Alpha (byte ordering, size_t etc).
232 - buffer overruns.
233 - use environment variable for TMPDIR (instead of /tmp).
234 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
235 - workaround egcs bug (RedHat bugs #4281 and #2989).
236 - wire $(OPT) throughout Makefile's.
237
238 2. Upgrade the dump revision to 1, making possible to dump filesystems
239 made with e2fsprogs-1.15 or newer. Nothing seems to break...
240
241 3. Fix some compile warnings, prototype all functions.
242
243 4. Use glibc err/glob instead of internal compatibility
244 routines (only if available).
245
246 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
247 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
248
249 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
250 ======================================================================
251
252 1. Integrated the changes from FreeBSD-3.1-RELEASE
253 (mostly bug fixes, buffer overruns, dump has now an "automatic
254 tape length calculation" flag, dump/restore can use kerberos now
255 (this is NOT tested), use environment variables for TAPE and
256 RMT etc.).
257
258 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
259
260 3. Corrected a bug in restore when using 'C' option with multi-volumes
261 tapes (files splited accros two tapes give "size changed" errors
262 when comparing).
263
264 4. Corrected the long standing bug when dumping multiple tapes.
265 This works for me, needs further testing.
266
267 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
268 ====================================================================
269
270 1. Dump now runs correctly on kernels 2.1.x
271 Fix made by Gerald Peters <gapeters@worldnet.att.net>
272
273 Changes between versions 0.4b2 and 0.4b3
274 ========================================
275
276 1. Use realpath() if available
277
278 2. Report statistics
279
280 Changes between versions 0.4b1 and 0.4b2
281 ========================================
282
283 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
284 the patch)
285
286 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
287
288 3. Fixed configure.in to correctly handle --enable-rmt
289
290 Changes between versions 0.3 and 0.4b1
291 ======================================
292
293 1. Integrated the changes from 4.4BSD-Lite2
294
295 2. Integrated the patches from Debian and Red Hat
296
297 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
298
299 4. Changed dump to use the Ext2fs library to get block addresses. This
300 should solve the endianness problem on SparcLinux.
301
302 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
303 distribution's one) to use autoconf
304
305 6. Fixed a few minor bugs
306
307 Changes between versions 0.2e and 0.2f
308 ======================================
309
310 1. Added the creation of named pipes (fifos) in restore.
311
312 2. Added the -N flag in the restore manual page.
313
314 3. Added the file kernel.patch which contains the llseek() optimization
315 patch for 1.2.x kernels.
316
317 4. Fixed a bug in the restoration of symbolic links: owner and group were
318 not restored.
319
320 5. Integrated some changes from FreeBSD 2.2.
321
322 6. Added a call to ftruncate() after restoring each file to restore
323 correctly files ending by a hole.
324
325 Changes between versions 0.2d and 0.2e
326 ======================================
327
328 1. Fixed a bug in the "set owner/mode" process. Every file was restored
329 with owner = root (0) and group = root/wheel/whatever (0).
330
331 Changes between versions 0.2c and 0.2d
332 ======================================
333
334 1. Dump is now able to backup 2GB+ filesystems.
335
336 2. Dump and restore can now be linked as static binaries.
337
338 Changes between versions 0.2b and 0.2c
339 ======================================
340
341 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
342
343 Changes between versions 0.2a and 0.2b
344 ======================================
345
346 1. Really fixed the bug that I should have corrected in 0.2a.
347
348 2. Enabled optimization again.
349
350 Changes between versions 0.2 and 0.2a
351 =====================================
352
353 1. Disabled the optimization during compilation.
354
355 Changes between versions 0.1 and 0.2
356 ====================================
357
358 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
359 the fs_type field (actually, I modified the file fstab.c to make it
360 use the mntent functions).
361
362 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
363 made by dump should be readable by the BSD restore and backups made
364 by the BSD dump should be readable by restore. Unfortunately, this
365 means that the dump format has changed between version 0.1 and version
366 0.2 :-(
367
368 3. Dump is now able to backup a subtree, it is no longer limited to whole
369 filesystems like the BSD version.
370
371 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
372 than 2 GB.
373
374 Changes between versions 0.0 and 0.1
375 ====================================
376
377 1. Now create links rdump and rrestore during the `make install' step.
378
379 2. Linux port specific bugs added to the manual pages
380
381 3. Incorrect estimation of the number of tapes blocks fixed when doing
382 incremental backups.
383
384 4. Better ls-like format in restore in interactive mode.