1 $Id: CHANGES,v 1.34 2000/01/11 12:32:53 tiniou Exp $
3 Changes between versions 0.4b12 and 0.4b13 (released ???????????????)
4 =====================================================================
6 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
7 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
9 2. Removed the requirement to build the RPM as root from the
10 spec file. Thanks to Christian Weisgerber
11 <naddy@mips.rhein-neckar.de> for submitting this.
13 3. Fix a bug in dump related to the 'filetype' feature of ext2fs,
14 causing dump to block when dumping really huge filesystems.
15 Many thanks to Patrik Schindler <poc@pocnet.net> for
16 helping me find this bug.
18 4. Fix the treatment for an interrupt signal when dump access
19 the remote tape through RSH. Thanks to Christian Weisgerber
20 <naddy@mips.rhein-neckar.de> for providing the patch.
22 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
23 =====================================================================
25 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
26 <kukuk@suse.de> for submitting the patch.
28 2. Fix for the exit code when using the size estimate option of
29 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
32 3. Handle EINTR in atomical reads/writes in dump, which was causing
33 dump to fail on some systems. Thanks to Eric Jergensen
34 <eric@dvns.com> for reporting the bug and submitting the patch.
36 4. Handle more than 16 characters for the device names in dumpdates.
37 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
38 tracking down the problem and proposing the solution.
40 5. Fixed a bug in dump which prevented the creation of the
41 dumpdates file when doing a 0-level dump without already
42 having a dumpdates file. Thanks to Patrik Schindler
43 <poc@pocnet.net> for reporting the bug.
45 6. Changed the way dump 'S' flag reports the size estimate
46 from number of blocks into bytes (making it compatible
47 with the Solaris version, and simplifying things for
48 amanda users). Thanks to Jason L Tibbitts III
49 <tibbs@math.uh.edu> for reporting the bug.
51 7. Fixed a compatibility problem in linux/alpha dump tape format.
52 Now the linux/alpha dump are (again) compatible with the
53 other dump formats. But this breaks compatibility with
54 older dumps made on alpha. Thanks to Mike Tibor
55 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
57 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
58 ======================================================================
60 1. Added a '--enable-kerberos' to configure.
62 2. Added a 'S' option to dump which determines the amount of space
63 that is needed to perform the dump without actually doing it, similar
64 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
65 <cermak@ahab.rutgers.edu>.
67 3. Added a 'M' multi-volume option to dump and restore which enables
68 dumping to multiple files (useful when dumping to an ext2
69 partition to make several dump files in order to bypass the 2GB
70 file size limitation). The 'f' argument is treated as a prefix and
71 the output files will be named <prefix>001, <prefix>002 etc. With
72 the 'M' flag, restore automatically selects the right file without
73 asking to enter a new tape each time.
75 4. Fixed a memory leak which was causing dump to grow very big
76 (270MB when dumping a 10GB filesystem...). Thanks to Jason
77 Fearon <jasonf@netrider.org.au> for reporting the bug.
79 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
80 ======================================================================
82 1. Make configure test if the system glob routines support
83 extended syntax (ALTDIRFUNC). If not, use the internal glob
84 routines instead of system ones. Thanks to Bernhard Sadlowski
85 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
86 and helping me resolve this and other minor libc5 compiling
89 2. Fix a problem when dumping a ext2fs with the 'filetype'
90 feature enabled. Thanks to Patrick J. LoPresti
91 <patl@cag.lcs.mit.edu> for reporting the bug and to
92 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
94 3. Made the nodump flag work on directories. A directory which
95 has the nodump flag gets never dumped, regardless of its
98 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
99 which allows dump on an active ext3 filesystem. However, this
100 is a "quick and dirty" patch which enables backup of an ext3
101 filesystem through the ext2 compatibility (by ignoring the
102 NEEDS_RECOVERY bit). The journal file is not recognized and
103 it is dumped (it should not).
105 5. Test the superblock compatibility flags when dumping, in order
106 to be sure that we know how to deal with specific features.
108 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
109 ====================================================================
111 1. Use lchown instead of chown, fixing a possible security problem
112 when restoring symlinks (a malicious user could use this
113 to deliberately corrupt the ownership of important system files).
114 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
115 this and providing the patch.
117 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
118 ====================================================================
120 1. Put dump sources under CVS, added Id tags in all files so
121 one can use 'ident' on binary files.
123 2. Added the dump/restore version in the usage text so one can
124 easily verify the version he is using.
126 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
127 a va_start/va_end problem on linux-ppc (always call va_start
128 va_end in pairs each time we use a vararg function).
130 4. Added again the DT_* constants because old libc does not
131 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
132 for submitting the bug report.
134 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
135 this should enable dumping big (huge) filesystems.
137 6. Added the RSH environment variable in order to be able to
138 use a rsh replacement like ssh when doing remote backups (and
139 bypass the security limitations of rcmd). Now you can do remote
140 backups without being root (or making dump setuid root).
142 7. Modified again the way dumpdates works. For incremental dumps,
143 we need to read dumpdates even if we are not using 'u' option.
144 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
147 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
148 ===================================================================
150 1. Removed the 'k' flag from the restore 'about' text if kerberos
153 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
154 to them (fsetflags takes a char*, setflags an open fd!).
156 3. (f)setflags is called only if the flags aren't empty. If the
157 file is a special file, a warning is printed, because changing
158 flags implies opening the device. Normally, a special file
159 should not have any flag... (Debian bug #29775, patch provided
160 by Abhijit Dasgupta <abhijit@ans.net>).
162 4. Made possible to dump a file system not mentioned in /etc/fstab.
163 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
165 5. Changed the default behaviour to not create dumpdates
166 unless 'u' option is specified. Removed the old "debian-patch"
167 which provided the same thing. (Debian bug #38136, #33818).
169 6. Removed all those dump*announce, since they were getting old...
171 7. Added warning messages if dumpdates does not exist and
172 when an inferior level dump does not exist (except for a level 0
175 8. Debugged the glob calls in interactive mode: restore used a
176 dirent struct which was different from the /usr/include/dirent.h
177 one (this used to work, is it a glibc2 change?), so none of the
178 compat glob (which used /usr/include/dirent.h) or the system glob
179 worked. Restore use now the system dirent (and the system
180 DT_* constants), which are compatible with BSD ones.
182 9. Added a configure flag (--with-dumpdatespath) to specify
183 the location of dumpdates file. By default, it is
186 10. Added the "AUTHOR" and "AVAILABILITY" sections and
187 included the current date/version in man pages.
189 11. Corrected the estimation of remaining time when
190 the operator doesn't change the tapes quickly enough. This
191 was an old bug, I thought I corrected it, and discovered
192 that in fact it was corrected in two different places, so
193 the results canceled each other...
195 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
196 ===================================================================
198 1. Integrated multiple patches from RedHat, Debian and SuSE:
200 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
202 - several fixes in the man pages.
203 - update the default tape device to /dev/st0.
204 - many updates for Linux Alpha (byte ordering, size_t etc).
206 - use environment variable for TMPDIR (instead of /tmp).
207 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
208 - workaround egcs bug (RedHat bugs #4281 and #2989).
209 - wire $(OPT) throughout Makefile's.
211 2. Upgrade the dump revision to 1, making possible to dump filesystems
212 made with e2fsprogs-1.15 or newer. Nothing seems to break...
214 3. Fix some compile warnings, prototype all functions.
216 4. Use glibc err/glob instead of internal compatibility
217 routines (only if available).
219 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
220 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
222 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
223 ======================================================================
225 1. Integrated the changes from FreeBSD-3.1-RELEASE
226 (mostly bug fixes, buffer overruns, dump has now an "automatic
227 tape length calculation" flag, dump/restore can use kerberos now
228 (this is NOT tested), use environment variables for TAPE and
231 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
233 3. Corrected a bug in restore when using 'C' option with multi-volumes
234 tapes (files splited accros two tapes give "size changed" errors
237 4. Corrected the long standing bug when dumping multiple tapes.
238 This works for me, needs further testing.
240 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
241 ====================================================================
243 1. Dump now runs correctly on kernels 2.1.x
244 Fix made by Gerald Peters <gapeters@worldnet.att.net>
246 Changes between versions 0.4b2 and 0.4b3
247 ========================================
249 1. Use realpath() if available
253 Changes between versions 0.4b1 and 0.4b2
254 ========================================
256 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
259 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
261 3. Fixed configure.in to correctly handle --enable-rmt
263 Changes between versions 0.3 and 0.4b1
264 ======================================
266 1. Integrated the changes from 4.4BSD-Lite2
268 2. Integrated the patches from Debian and Red Hat
270 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
272 4. Changed dump to use the Ext2fs library to get block addresses. This
273 should solve the endianness problem on SparcLinux.
275 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
276 distribution's one) to use autoconf
278 6. Fixed a few minor bugs
280 Changes between versions 0.2e and 0.2f
281 ======================================
283 1. Added the creation of named pipes (fifos) in restore.
285 2. Added the -N flag in the restore manual page.
287 3. Added the file kernel.patch which contains the llseek() optimization
288 patch for 1.2.x kernels.
290 4. Fixed a bug in the restoration of symbolic links: owner and group were
293 5. Integrated some changes from FreeBSD 2.2.
295 6. Added a call to ftruncate() after restoring each file to restore
296 correctly files ending by a hole.
298 Changes between versions 0.2d and 0.2e
299 ======================================
301 1. Fixed a bug in the "set owner/mode" process. Every file was restored
302 with owner = root (0) and group = root/wheel/whatever (0).
304 Changes between versions 0.2c and 0.2d
305 ======================================
307 1. Dump is now able to backup 2GB+ filesystems.
309 2. Dump and restore can now be linked as static binaries.
311 Changes between versions 0.2b and 0.2c
312 ======================================
314 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
316 Changes between versions 0.2a and 0.2b
317 ======================================
319 1. Really fixed the bug that I should have corrected in 0.2a.
321 2. Enabled optimization again.
323 Changes between versions 0.2 and 0.2a
324 =====================================
326 1. Disabled the optimization during compilation.
328 Changes between versions 0.1 and 0.2
329 ====================================
331 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
332 the fs_type field (actually, I modified the file fstab.c to make it
333 use the mntent functions).
335 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
336 made by dump should be readable by the BSD restore and backups made
337 by the BSD dump should be readable by restore. Unfortunately, this
338 means that the dump format has changed between version 0.1 and version
341 3. Dump is now able to backup a subtree, it is no longer limited to whole
342 filesystems like the BSD version.
344 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
347 Changes between versions 0.0 and 0.1
348 ====================================
350 1. Now create links rdump and rrestore during the `make install' step.
352 2. Linux port specific bugs added to the manual pages
354 3. Incorrect estimation of the number of tapes blocks fixed when doing
357 4. Better ls-like format in restore in interactive mode.