]> git.wh0rd.org - dump.git/blob - CHANGES
Dumping a mountpoint caused the mountpoint to be dumped instead of the mounted disk.
[dump.git] / CHANGES
1 $Id: CHANGES,v 1.38 2000/01/26 11:38:07 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 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
22 ======================================================================
23
24 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
25 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
26
27 2. Removed the requirement to build the RPM as root from the
28 spec file. Thanks to Christian Weisgerber
29 <naddy@mips.rhein-neckar.de> for submitting this.
30
31 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
32 causing dump to block when dumping really huge filesystems.
33 Many thanks to Patrik Schindler <poc@pocnet.net> for
34 helping me find this bug.
35
36 4. Fixed the treatment for an interrupt signal when dump access
37 the remote tape through RSH. Thanks to Christian Weisgerber
38 <naddy@mips.rhein-neckar.de> for providing the patch.
39
40 5. Fixed a bug which was causing dump/restore to display
41 garbage characters instead of the remote host name.
42
43 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
44 =====================================================================
45
46 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
47 <kukuk@suse.de> for submitting the patch.
48
49 2. Fix for the exit code when using the size estimate option of
50 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
51 submitting the patch.
52
53 3. Handle EINTR in atomical reads/writes in dump, which was causing
54 dump to fail on some systems. Thanks to Eric Jergensen
55 <eric@dvns.com> for reporting the bug and submitting the patch.
56
57 4. Handle more than 16 characters for the device names in dumpdates.
58 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
59 tracking down the problem and proposing the solution.
60
61 5. Fixed a bug in dump which prevented the creation of the
62 dumpdates file when doing a 0-level dump without already
63 having a dumpdates file. Thanks to Patrik Schindler
64 <poc@pocnet.net> for reporting the bug.
65
66 6. Changed the way dump 'S' flag reports the size estimate
67 from number of blocks into bytes (making it compatible
68 with the Solaris version, and simplifying things for
69 amanda users). Thanks to Jason L Tibbitts III
70 <tibbs@math.uh.edu> for reporting the bug.
71
72 7. Fixed a compatibility problem in linux/alpha dump tape format.
73 Now the linux/alpha dump are (again) compatible with the
74 other dump formats. But this breaks compatibility with
75 older dumps made on alpha. Thanks to Mike Tibor
76 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
77
78 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
79 ======================================================================
80
81 1. Added a '--enable-kerberos' to configure.
82
83 2. Added a 'S' option to dump which determines the amount of space
84 that is needed to perform the dump without actually doing it, similar
85 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
86 <cermak@ahab.rutgers.edu>.
87
88 3. Added a 'M' multi-volume option to dump and restore which enables
89 dumping to multiple files (useful when dumping to an ext2
90 partition to make several dump files in order to bypass the 2GB
91 file size limitation). The 'f' argument is treated as a prefix and
92 the output files will be named <prefix>001, <prefix>002 etc. With
93 the 'M' flag, restore automatically selects the right file without
94 asking to enter a new tape each time.
95
96 4. Fixed a memory leak which was causing dump to grow very big
97 (270MB when dumping a 10GB filesystem...). Thanks to Jason
98 Fearon <jasonf@netrider.org.au> for reporting the bug.
99
100 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
101 ======================================================================
102
103 1. Make configure test if the system glob routines support
104 extended syntax (ALTDIRFUNC). If not, use the internal glob
105 routines instead of system ones. Thanks to Bernhard Sadlowski
106 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
107 and helping me resolve this and other minor libc5 compiling
108 glitches.
109
110 2. Fix a problem when dumping a ext2fs with the 'filetype'
111 feature enabled. Thanks to Patrick J. LoPresti
112 <patl@cag.lcs.mit.edu> for reporting the bug and to
113 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
114
115 3. Made the nodump flag work on directories. A directory which
116 has the nodump flag gets never dumped, regardless of its
117 contents.
118
119 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
120 which allows dump on an active ext3 filesystem. However, this
121 is a "quick and dirty" patch which enables backup of an ext3
122 filesystem through the ext2 compatibility (by ignoring the
123 NEEDS_RECOVERY bit). The journal file is not recognized and
124 it is dumped (it should not).
125
126 5. Test the superblock compatibility flags when dumping, in order
127 to be sure that we know how to deal with specific features.
128
129 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
130 ====================================================================
131
132 1. Use lchown instead of chown, fixing a possible security problem
133 when restoring symlinks (a malicious user could use this
134 to deliberately corrupt the ownership of important system files).
135 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
136 this and providing the patch.
137
138 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
139 ====================================================================
140
141 1. Put dump sources under CVS, added Id tags in all files so
142 one can use 'ident' on binary files.
143
144 2. Added the dump/restore version in the usage text so one can
145 easily verify the version he is using.
146
147 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
148 a va_start/va_end problem on linux-ppc (always call va_start
149 va_end in pairs each time we use a vararg function).
150
151 4. Added again the DT_* constants because old libc does not
152 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
153 for submitting the bug report.
154
155 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
156 this should enable dumping big (huge) filesystems.
157
158 6. Added the RSH environment variable in order to be able to
159 use a rsh replacement like ssh when doing remote backups (and
160 bypass the security limitations of rcmd). Now you can do remote
161 backups without being root (or making dump setuid root).
162
163 7. Modified again the way dumpdates works. For incremental dumps,
164 we need to read dumpdates even if we are not using 'u' option.
165 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
166 this should work.
167
168 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
169 ===================================================================
170
171 1. Removed the 'k' flag from the restore 'about' text if kerberos
172 was not compiled in.
173
174 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
175 to them (fsetflags takes a char*, setflags an open fd!).
176
177 3. (f)setflags is called only if the flags aren't empty. If the
178 file is a special file, a warning is printed, because changing
179 flags implies opening the device. Normally, a special file
180 should not have any flag... (Debian bug #29775, patch provided
181 by Abhijit Dasgupta <abhijit@ans.net>).
182
183 4. Made possible to dump a file system not mentioned in /etc/fstab.
184 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
185
186 5. Changed the default behaviour to not create dumpdates
187 unless 'u' option is specified. Removed the old "debian-patch"
188 which provided the same thing. (Debian bug #38136, #33818).
189
190 6. Removed all those dump*announce, since they were getting old...
191
192 7. Added warning messages if dumpdates does not exist and
193 when an inferior level dump does not exist (except for a level 0
194 dump).
195
196 8. Debugged the glob calls in interactive mode: restore used a
197 dirent struct which was different from the /usr/include/dirent.h
198 one (this used to work, is it a glibc2 change?), so none of the
199 compat glob (which used /usr/include/dirent.h) or the system glob
200 worked. Restore use now the system dirent (and the system
201 DT_* constants), which are compatible with BSD ones.
202
203 9. Added a configure flag (--with-dumpdatespath) to specify
204 the location of dumpdates file. By default, it is
205 /etc/dumpdates.
206
207 10. Added the "AUTHOR" and "AVAILABILITY" sections and
208 included the current date/version in man pages.
209
210 11. Corrected the estimation of remaining time when
211 the operator doesn't change the tapes quickly enough. This
212 was an old bug, I thought I corrected it, and discovered
213 that in fact it was corrected in two different places, so
214 the results canceled each other...
215
216 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
217 ===================================================================
218
219 1. Integrated multiple patches from RedHat, Debian and SuSE:
220
221 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
222 isn't specified.
223 - several fixes in the man pages.
224 - update the default tape device to /dev/st0.
225 - many updates for Linux Alpha (byte ordering, size_t etc).
226 - buffer overruns.
227 - use environment variable for TMPDIR (instead of /tmp).
228 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
229 - workaround egcs bug (RedHat bugs #4281 and #2989).
230 - wire $(OPT) throughout Makefile's.
231
232 2. Upgrade the dump revision to 1, making possible to dump filesystems
233 made with e2fsprogs-1.15 or newer. Nothing seems to break...
234
235 3. Fix some compile warnings, prototype all functions.
236
237 4. Use glibc err/glob instead of internal compatibility
238 routines (only if available).
239
240 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
241 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
242
243 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
244 ======================================================================
245
246 1. Integrated the changes from FreeBSD-3.1-RELEASE
247 (mostly bug fixes, buffer overruns, dump has now an "automatic
248 tape length calculation" flag, dump/restore can use kerberos now
249 (this is NOT tested), use environment variables for TAPE and
250 RMT etc.).
251
252 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
253
254 3. Corrected a bug in restore when using 'C' option with multi-volumes
255 tapes (files splited accros two tapes give "size changed" errors
256 when comparing).
257
258 4. Corrected the long standing bug when dumping multiple tapes.
259 This works for me, needs further testing.
260
261 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
262 ====================================================================
263
264 1. Dump now runs correctly on kernels 2.1.x
265 Fix made by Gerald Peters <gapeters@worldnet.att.net>
266
267 Changes between versions 0.4b2 and 0.4b3
268 ========================================
269
270 1. Use realpath() if available
271
272 2. Report statistics
273
274 Changes between versions 0.4b1 and 0.4b2
275 ========================================
276
277 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
278 the patch)
279
280 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
281
282 3. Fixed configure.in to correctly handle --enable-rmt
283
284 Changes between versions 0.3 and 0.4b1
285 ======================================
286
287 1. Integrated the changes from 4.4BSD-Lite2
288
289 2. Integrated the patches from Debian and Red Hat
290
291 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
292
293 4. Changed dump to use the Ext2fs library to get block addresses. This
294 should solve the endianness problem on SparcLinux.
295
296 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
297 distribution's one) to use autoconf
298
299 6. Fixed a few minor bugs
300
301 Changes between versions 0.2e and 0.2f
302 ======================================
303
304 1. Added the creation of named pipes (fifos) in restore.
305
306 2. Added the -N flag in the restore manual page.
307
308 3. Added the file kernel.patch which contains the llseek() optimization
309 patch for 1.2.x kernels.
310
311 4. Fixed a bug in the restoration of symbolic links: owner and group were
312 not restored.
313
314 5. Integrated some changes from FreeBSD 2.2.
315
316 6. Added a call to ftruncate() after restoring each file to restore
317 correctly files ending by a hole.
318
319 Changes between versions 0.2d and 0.2e
320 ======================================
321
322 1. Fixed a bug in the "set owner/mode" process. Every file was restored
323 with owner = root (0) and group = root/wheel/whatever (0).
324
325 Changes between versions 0.2c and 0.2d
326 ======================================
327
328 1. Dump is now able to backup 2GB+ filesystems.
329
330 2. Dump and restore can now be linked as static binaries.
331
332 Changes between versions 0.2b and 0.2c
333 ======================================
334
335 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
336
337 Changes between versions 0.2a and 0.2b
338 ======================================
339
340 1. Really fixed the bug that I should have corrected in 0.2a.
341
342 2. Enabled optimization again.
343
344 Changes between versions 0.2 and 0.2a
345 =====================================
346
347 1. Disabled the optimization during compilation.
348
349 Changes between versions 0.1 and 0.2
350 ====================================
351
352 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
353 the fs_type field (actually, I modified the file fstab.c to make it
354 use the mntent functions).
355
356 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
357 made by dump should be readable by the BSD restore and backups made
358 by the BSD dump should be readable by restore. Unfortunately, this
359 means that the dump format has changed between version 0.1 and version
360 0.2 :-(
361
362 3. Dump is now able to backup a subtree, it is no longer limited to whole
363 filesystems like the BSD version.
364
365 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
366 than 2 GB.
367
368 Changes between versions 0.0 and 0.1
369 ====================================
370
371 1. Now create links rdump and rrestore during the `make install' step.
372
373 2. Linux port specific bugs added to the manual pages
374
375 3. Incorrect estimation of the number of tapes blocks fixed when doing
376 incremental backups.
377
378 4. Better ls-like format in restore in interactive mode.