]>
Commit | Line | Data |
---|---|---|
d3393043 SP |
1 | $Id: CHANGES,v 1.44 2000/02/26 01:35:48 stelian Exp $ |
2 | ||
3 | Changes between versions 0.4b14 and 0.4b15 (released ?????????????????) | |
4 | ======================================================================= | |
5 | ||
6 | 1. Added a prompt command in interactive restore mode. Thanks | |
7 | to Andreas Dilger <adilger@home.com> for the patch. | |
d6f78b13 | 8 | |
688653ea SP |
9 | Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000) |
10 | ======================================================================= | |
d6f78b13 SP |
11 | |
12 | 1. Fixed a bug in dump which may have caused invalid deleted | |
13 | directories to be dumped out if they were deleted after the | |
14 | mapping pass. This could occure on active filesystem and lead | |
15 | to heap corruption (causing dump malfunction in many possible ways). | |
16 | Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this | |
17 | bug and submitting the patch. | |
ae58c8da | 18 | |
a2c9bd28 SP |
19 | 2. Fixed the handling of the filesystem parameter in dump. A |
20 | '/mnt/disk' parameter caused the disk contents to be dumped, | |
21 | but a '/mnt/disk/' parameter caused the mountpoint directory | |
22 | to be dumped (generally an empty directory). | |
23 | ||
24 | 3. Improved the output of dump in order to tell which directory | |
25 | it is currently dumping (when dumping a subtree). | |
26 | ||
20c345aa SP |
27 | 4. Added the '-e' exclude inode option to dump. Thanks to |
28 | Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch. | |
29 | ||
d8476fa5 SP |
30 | 5. Added a REPORTING-BUGS file in order to provide a guide |
31 | on how to correctly report a bug in dump/restore. | |
32 | ||
c60338a5 SP |
33 | 6. Fixed a restore problem when restoring a hard link to an inode |
34 | having the immutable or append-only attribute set. Thanks to | |
35 | Ambrose Li <acli@mingpaoxpress.com> for submitting the patch. | |
36 | ||
8eb05297 SP |
37 | 7. Fixed a compatibility problem between dumps made on little |
38 | endian machines (the format was correct) and big endian | |
39 | machines (the format was incorrect). This fix break the | |
40 | compatibility with the older dumps made on big endian | |
41 | machines (sparc, m86k, ppc etc). For the first time in | |
42 | linux dump's history (I believe), the dumps made by *BSD, | |
43 | Linux/alpha, Linux/sparc and Linux/x86 are compatible, | |
44 | so interchangeable. Thanks to Rob Cermak | |
45 | <cermak@ahab.rutgers.edu> for submitting the bug and | |
46 | helping me test the fix. | |
47 | ||
48 | 8. Fixed the way dump reports the remaining percent/time, if | |
49 | the number of blocks actually dumped exceeds the estimated | |
50 | number of blocks. Thanks to Jean-Paul van der Jagt | |
51 | <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug. | |
52 | ||
d4b74b5c SP |
53 | Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000) |
54 | ====================================================================== | |
2f9a0c5c SP |
55 | |
56 | 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski | |
57 | <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug. | |
58 | ||
6dda05e4 SP |
59 | 2. Removed the requirement to build the RPM as root from the |
60 | spec file. Thanks to Christian Weisgerber | |
61 | <naddy@mips.rhein-neckar.de> for submitting this. | |
62 | ||
8d957ae7 | 63 | 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs, |
f346e363 SP |
64 | causing dump to block when dumping really huge filesystems. |
65 | Many thanks to Patrik Schindler <poc@pocnet.net> for | |
66 | helping me find this bug. | |
67 | ||
8d957ae7 | 68 | 4. Fixed the treatment for an interrupt signal when dump access |
b80c25f8 SP |
69 | the remote tape through RSH. Thanks to Christian Weisgerber |
70 | <naddy@mips.rhein-neckar.de> for providing the patch. | |
71 | ||
8d957ae7 SP |
72 | 5. Fixed a bug which was causing dump/restore to display |
73 | garbage characters instead of the remote host name. | |
74 | ||
2f9a0c5c | 75 | Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000) |
df172bbb | 76 | ===================================================================== |
ae58c8da SP |
77 | |
78 | 1. Small fix in the dump man page. Thanks to Thorsten Kukuk | |
79 | <kukuk@suse.de> for submitting the patch. | |
2ed89fda | 80 | |
a29c23cc SP |
81 | 2. Fix for the exit code when using the size estimate option of |
82 | dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for | |
83 | submitting the patch. | |
84 | ||
2e682895 SP |
85 | 3. Handle EINTR in atomical reads/writes in dump, which was causing |
86 | dump to fail on some systems. Thanks to Eric Jergensen | |
87 | <eric@dvns.com> for reporting the bug and submitting the patch. | |
88 | ||
89 | 4. Handle more than 16 characters for the device names in dumpdates. | |
90 | (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for | |
91 | tracking down the problem and proposing the solution. | |
92 | ||
93 | 5. Fixed a bug in dump which prevented the creation of the | |
94 | dumpdates file when doing a 0-level dump without already | |
95 | having a dumpdates file. Thanks to Patrik Schindler | |
96 | <poc@pocnet.net> for reporting the bug. | |
97 | ||
98 | 6. Changed the way dump 'S' flag reports the size estimate | |
99 | from number of blocks into bytes (making it compatible | |
100 | with the Solaris version, and simplifying things for | |
101 | amanda users). Thanks to Jason L Tibbitts III | |
102 | <tibbs@math.uh.edu> for reporting the bug. | |
103 | ||
0d0bc326 SP |
104 | 7. Fixed a compatibility problem in linux/alpha dump tape format. |
105 | Now the linux/alpha dump are (again) compatible with the | |
106 | other dump formats. But this breaks compatibility with | |
107 | older dumps made on alpha. Thanks to Mike Tibor | |
108 | <tibor@lib.uaa.alaska.edu> for helping me in finding this bug. | |
109 | ||
cf9aab8e | 110 | Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999) |
2ed89fda SP |
111 | ====================================================================== |
112 | ||
113 | 1. Added a '--enable-kerberos' to configure. | |
5c1faba5 | 114 | |
144a6db1 | 115 | 2. Added a 'S' option to dump which determines the amount of space |
5c729713 | 116 | that is needed to perform the dump without actually doing it, similar |
144a6db1 SP |
117 | to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak |
118 | <cermak@ahab.rutgers.edu>. | |
119 | ||
677ee424 SP |
120 | 3. Added a 'M' multi-volume option to dump and restore which enables |
121 | dumping to multiple files (useful when dumping to an ext2 | |
122 | partition to make several dump files in order to bypass the 2GB | |
123 | file size limitation). The 'f' argument is treated as a prefix and | |
124 | the output files will be named <prefix>001, <prefix>002 etc. With | |
125 | the 'M' flag, restore automatically selects the right file without | |
126 | asking to enter a new tape each time. | |
127 | ||
7fbb9b02 | 128 | 4. Fixed a memory leak which was causing dump to grow very big |
a29c23cc SP |
129 | (270MB when dumping a 10GB filesystem...). Thanks to Jason |
130 | Fearon <jasonf@netrider.org.au> for reporting the bug. | |
7fbb9b02 | 131 | |
27026010 SP |
132 | Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999) |
133 | ====================================================================== | |
5c1faba5 SP |
134 | |
135 | 1. Make configure test if the system glob routines support | |
136 | extended syntax (ALTDIRFUNC). If not, use the internal glob | |
a95c4c36 SP |
137 | routines instead of system ones. Thanks to Bernhard Sadlowski |
138 | <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug | |
139 | and helping me resolve this and other minor libc5 compiling | |
140 | glitches. | |
08ebf8d7 | 141 | |
c5886bd7 SP |
142 | 2. Fix a problem when dumping a ext2fs with the 'filetype' |
143 | feature enabled. Thanks to Patrick J. LoPresti | |
144 | <patl@cag.lcs.mit.edu> for reporting the bug and to | |
145 | Theodore Y. Ts'o <tytso@mit.edu> for providing the patch. | |
146 | ||
4267a8ad SP |
147 | 3. Made the nodump flag work on directories. A directory which |
148 | has the nodump flag gets never dumped, regardless of its | |
149 | contents. | |
150 | ||
c930abff SP |
151 | 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org> |
152 | which allows dump on an active ext3 filesystem. However, this | |
153 | is a "quick and dirty" patch which enables backup of an ext3 | |
154 | filesystem through the ext2 compatibility (by ignoring the | |
155 | NEEDS_RECOVERY bit). The journal file is not recognized and | |
156 | it is dumped (it should not). | |
157 | ||
158 | 5. Test the superblock compatibility flags when dumping, in order | |
159 | to be sure that we know how to deal with specific features. | |
160 | ||
08ebf8d7 SP |
161 | Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999) |
162 | ==================================================================== | |
163 | ||
164 | 1. Use lchown instead of chown, fixing a possible security problem | |
165 | when restoring symlinks (a malicious user could use this | |
166 | to deliberately corrupt the ownership of important system files). | |
167 | Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting | |
168 | this and providing the patch. | |
ec387a12 | 169 | |
c154548c SP |
170 | Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999) |
171 | ==================================================================== | |
ec387a12 | 172 | |
df9ae507 SP |
173 | 1. Put dump sources under CVS, added Id tags in all files so |
174 | one can use 'ident' on binary files. | |
175 | ||
176 | 2. Added the dump/restore version in the usage text so one can | |
177 | easily verify the version he is using. | |
ec387a12 | 178 | |
3072400e SP |
179 | 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes |
180 | a va_start/va_end problem on linux-ppc (always call va_start | |
181 | va_end in pairs each time we use a vararg function). | |
182 | ||
fa58b58d SP |
183 | 4. Added again the DT_* constants because old libc does not |
184 | contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr> | |
185 | for submitting the bug report. | |
186 | ||
f34aca37 SP |
187 | 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs |
188 | this should enable dumping big (huge) filesystems. | |
189 | ||
190 | 6. Added the RSH environment variable in order to be able to | |
0c62667d SP |
191 | use a rsh replacement like ssh when doing remote backups (and |
192 | bypass the security limitations of rcmd). Now you can do remote | |
193 | backups without being root (or making dump setuid root). | |
194 | ||
f34aca37 | 195 | 7. Modified again the way dumpdates works. For incremental dumps, |
60f5ede9 SP |
196 | we need to read dumpdates even if we are not using 'u' option. |
197 | Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how | |
198 | this should work. | |
199 | ||
8d4197bb SP |
200 | Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999) |
201 | =================================================================== | |
202 | ||
203 | 1. Removed the 'k' flag from the restore 'about' text if kerberos | |
204 | was not compiled in. | |
205 | ||
206 | 2. Prototyped (f)setflags from e2fsprogs and corrected the calls | |
207 | to them (fsetflags takes a char*, setflags an open fd!). | |
208 | ||
209 | 3. (f)setflags is called only if the flags aren't empty. If the | |
210 | file is a special file, a warning is printed, because changing | |
211 | flags implies opening the device. Normally, a special file | |
212 | should not have any flag... (Debian bug #29775, patch provided | |
213 | by Abhijit Dasgupta <abhijit@ans.net>). | |
214 | ||
215 | 4. Made possible to dump a file system not mentioned in /etc/fstab. | |
216 | (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>). | |
217 | ||
218 | 5. Changed the default behaviour to not create dumpdates | |
219 | unless 'u' option is specified. Removed the old "debian-patch" | |
220 | which provided the same thing. (Debian bug #38136, #33818). | |
221 | ||
222 | 6. Removed all those dump*announce, since they were getting old... | |
223 | ||
224 | 7. Added warning messages if dumpdates does not exist and | |
225 | when an inferior level dump does not exist (except for a level 0 | |
226 | dump). | |
227 | ||
228 | 8. Debugged the glob calls in interactive mode: restore used a | |
229 | dirent struct which was different from the /usr/include/dirent.h | |
230 | one (this used to work, is it a glibc2 change?), so none of the | |
231 | compat glob (which used /usr/include/dirent.h) or the system glob | |
232 | worked. Restore use now the system dirent (and the system | |
233 | DT_* constants), which are compatible with BSD ones. | |
234 | ||
235 | 9. Added a configure flag (--with-dumpdatespath) to specify | |
236 | the location of dumpdates file. By default, it is | |
237 | /etc/dumpdates. | |
238 | ||
239 | 10. Added the "AUTHOR" and "AVAILABILITY" sections and | |
240 | included the current date/version in man pages. | |
241 | ||
242 | 11. Corrected the estimation of remaining time when | |
243 | the operator doesn't change the tapes quickly enough. This | |
244 | was an old bug, I thought I corrected it, and discovered | |
245 | that in fact it was corrected in two different places, so | |
246 | the results canceled each other... | |
247 | ||
248 | Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999) | |
249 | =================================================================== | |
ddd2ef55 SP |
250 | |
251 | 1. Integrated multiple patches from RedHat, Debian and SuSE: | |
252 | ||
253 | - tweak dump/itime.c to not try to read dumpdates if the 'u' option | |
254 | isn't specified. | |
255 | - several fixes in the man pages. | |
256 | - update the default tape device to /dev/st0. | |
257 | - many updates for Linux Alpha (byte ordering, size_t etc). | |
258 | - buffer overruns. | |
259 | - use environment variable for TMPDIR (instead of /tmp). | |
260 | - use sigjmp_buf instead of jmp_buf (RedHat bug #3260). | |
261 | - workaround egcs bug (RedHat bugs #4281 and #2989). | |
262 | - wire $(OPT) throughout Makefile's. | |
263 | ||
8d4197bb SP |
264 | 2. Upgrade the dump revision to 1, making possible to dump filesystems |
265 | made with e2fsprogs-1.15 or newer. Nothing seems to break... | |
ddd2ef55 | 266 | |
8d4197bb SP |
267 | 3. Fix some compile warnings, prototype all functions. |
268 | ||
269 | 4. Use glibc err/glob instead of internal compatibility | |
ddd2ef55 SP |
270 | routines (only if available). |
271 | ||
8d4197bb SP |
272 | 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided |
273 | by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>). | |
274 | ||
275 | Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999) | |
276 | ====================================================================== | |
b45f51d6 SP |
277 | |
278 | 1. Integrated the changes from FreeBSD-3.1-RELEASE | |
279 | (mostly bug fixes, buffer overruns, dump has now an "automatic | |
280 | tape length calculation" flag, dump/restore can use kerberos now | |
281 | (this is NOT tested), use environment variables for TAPE and | |
282 | RMT etc.). | |
283 | ||
284 | 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches) | |
285 | ||
286 | 3. Corrected a bug in restore when using 'C' option with multi-volumes | |
287 | tapes (files splited accros two tapes give "size changed" errors | |
288 | when comparing). | |
289 | ||
290 | 4. Corrected the long standing bug when dumping multiple tapes. | |
291 | This works for me, needs further testing. | |
1227625a | 292 | |
8d4197bb SP |
293 | Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997) |
294 | ==================================================================== | |
1227625a SP |
295 | |
296 | 1. Dump now runs correctly on kernels 2.1.x | |
297 | Fix made by Gerald Peters <gapeters@worldnet.att.net> | |
298 | ||
299 | Changes between versions 0.4b2 and 0.4b3 | |
300 | ======================================== | |
301 | ||
302 | 1. Use realpath() if available | |
303 | ||
304 | 2. Report statistics | |
305 | ||
306 | Changes between versions 0.4b1 and 0.4b2 | |
307 | ======================================== | |
308 | ||
309 | 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating | |
310 | the patch) | |
311 | ||
312 | 2. Fixed restore to make it able to read FreeBSD 2.x dumps again | |
313 | ||
314 | 3. Fixed configure.in to correctly handle --enable-rmt | |
315 | ||
316 | Changes between versions 0.3 and 0.4b1 | |
317 | ====================================== | |
318 | ||
319 | 1. Integrated the changes from 4.4BSD-Lite2 | |
320 | ||
321 | 2. Integrated the patches from Debian and Red Hat | |
322 | ||
323 | 3. Portability changes: use the __u32, __u16, __s32, and __s16 types | |
324 | ||
325 | 4. Changed dump to use the Ext2fs library to get block addresses. This | |
326 | should solve the endianness problem on SparcLinux. | |
327 | ||
328 | 5. Created a configure.in file (shamelessly stolen from the e2fsprogs | |
329 | distribution's one) to use autoconf | |
330 | ||
331 | 6. Fixed a few minor bugs | |
332 | ||
333 | Changes between versions 0.2e and 0.2f | |
334 | ====================================== | |
335 | ||
336 | 1. Added the creation of named pipes (fifos) in restore. | |
337 | ||
338 | 2. Added the -N flag in the restore manual page. | |
339 | ||
340 | 3. Added the file kernel.patch which contains the llseek() optimization | |
341 | patch for 1.2.x kernels. | |
342 | ||
343 | 4. Fixed a bug in the restoration of symbolic links: owner and group were | |
344 | not restored. | |
345 | ||
346 | 5. Integrated some changes from FreeBSD 2.2. | |
347 | ||
348 | 6. Added a call to ftruncate() after restoring each file to restore | |
349 | correctly files ending by a hole. | |
350 | ||
351 | Changes between versions 0.2d and 0.2e | |
352 | ====================================== | |
353 | ||
354 | 1. Fixed a bug in the "set owner/mode" process. Every file was restored | |
355 | with owner = root (0) and group = root/wheel/whatever (0). | |
356 | ||
357 | Changes between versions 0.2c and 0.2d | |
358 | ====================================== | |
359 | ||
360 | 1. Dump is now able to backup 2GB+ filesystems. | |
361 | ||
362 | 2. Dump and restore can now be linked as static binaries. | |
363 | ||
364 | Changes between versions 0.2b and 0.2c | |
365 | ====================================== | |
366 | ||
367 | 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links. | |
368 | ||
369 | Changes between versions 0.2a and 0.2b | |
370 | ====================================== | |
371 | ||
372 | 1. Really fixed the bug that I should have corrected in 0.2a. | |
373 | ||
374 | 2. Enabled optimization again. | |
375 | ||
376 | Changes between versions 0.2 and 0.2a | |
377 | ===================================== | |
378 | ||
379 | 1. Disabled the optimization during compilation. | |
380 | ||
381 | Changes between versions 0.1 and 0.2 | |
382 | ==================================== | |
383 | ||
384 | 1. Fixed a bug in fstab.c which caused a null pointer to be stored in | |
385 | the fs_type field (actually, I modified the file fstab.c to make it | |
386 | use the mntent functions). | |
387 | ||
388 | 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups | |
389 | made by dump should be readable by the BSD restore and backups made | |
390 | by the BSD dump should be readable by restore. Unfortunately, this | |
391 | means that the dump format has changed between version 0.1 and version | |
392 | 0.2 :-( | |
393 | ||
394 | 3. Dump is now able to backup a subtree, it is no longer limited to whole | |
395 | filesystems like the BSD version. | |
396 | ||
397 | 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger | |
398 | than 2 GB. | |
399 | ||
400 | Changes between versions 0.0 and 0.1 | |
401 | ==================================== | |
402 | ||
403 | 1. Now create links rdump and rrestore during the `make install' step. | |
404 | ||
405 | 2. Linux port specific bugs added to the manual pages | |
406 | ||
407 | 3. Incorrect estimation of the number of tapes blocks fixed when doing | |
408 | incremental backups. | |
409 | ||
410 | 4. Better ls-like format in restore in interactive mode. |