-$Id: CHANGES,v 1.54 2000/03/11 15:34:54 stelian Exp $
+$Id: CHANGES,v 1.120 2001/07/18 09:50:48 stelian Exp $
+
+Changes between versions 0.4b22 and 0.4b23 (released ????????????)
+==================================================================
+
+1. Fixed a buffer overflow in restore/tape.c. Patch provided by
+ Marcus Meissner (Caldera International Security Dept.).
+
+2. Implement the Sun rmt extended protocol. Patch submitted
+ by Ian Gordon <iangordon@users.sourceforge.net>.
+
+3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
+ the linux kernel header. This ensures that dump always has the
+ latest version of this file. Patch submitted by
+ Andreas Dilger <adilger@turbolinux.com>.
+
+4. Report any filesystem present in either /etc/fstab with a
+ positive passno or /etc/dumpdates in dump -w output.
+ Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
+
+5. Fixed the looping problem in dump introduced in the
+ previous version.
+
+6. Changed the -B option of dump to limit the size of
+ _compressed_ output per volume if compression is on.
+ Patch contributed by Helmut Jarausch
+ <jarausch@igpm.rwth-aachen.de>.
+
+Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
+==================================================================
+
+1. Made dump report the number of blocks dumped per volume.
+ Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
+
+2. Fix a bug in dump where waiting too much at the 'change volumes'
+ question caused the volume to be rewritten. Thanks to
+ Graham Reed <greed@users.sourceforge.net> for reporting the
+ bug and providing a patch.
+
+3. Added a compression option to dump, contributed by Jerry
+ Peters <gapeters@worldnet.att.net>.
+
+ WARNING: the compression works only on files/pipes and
+ drives supporting variable block size.
+
+ WARNING2: tapes created using the compression feature are
+ incompatible with the standard dump tape format, and a
+ version of restore >= 0.4b22 will be needed for extraction.
+
+4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
+ architectures. Thanks to Paul Slootman <paul@debian.org> for
+ the patch and to Bdale Garbee <bdale@gag.com> for forwarding
+ it upstream.
+
+5. Many cleanups (CPP defines, const char warnings, check of
+ ext2fs COMPAT flags, time_t cleanups, added libext2 version
+ in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
+
+6. Made --prefix option work in configure. All the install path
+ are now based on the configure parameters.
+
+7. Added the Quick File Access mode in dump/restore, contributed
+ by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
+ a file tape position for each inode, and this file is used by
+ restore (if called with parameter Q and the filename)
+ to directly position the tape at the file restore is currently
+ working on. This saves hours when restoring single files from
+ large backups, saves the tapes and the drive's head. Use
+ --enable-qfa option of configure to compile in the QFA support.
+
+8. Added the possibility to dump several files and directories
+ in a single invocation of dump. Thanks to Uwe Gohlke
+ <uwe@ugsoft.de> for implementing this option.
+
+9. Fixed the dumping and restoring of files with holes
+ (files ending with holes didn't get dumped or restored
+ correctly).
+
+10. Fixed a socket descriptor leak in dump, which leaved opened
+ 3 file descriptors per dump process (and there is one dump
+ process per tape).
+
+11. Fixed dump large file system support, by explicit use of
+ open64/lseek64/etc functions (explicit use needed because
+ e2fsprogs libraries don't behave well when compiled with
+ FILE_OFFSET_BITS=64).
+
+Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
+======================================================================
+
+1. Fixed some bugs in the dump -w|-W logic introduced by
+ the previous version. Thanks to Andreas Dilger
+ <adilger@turbolinux.com> for his help on this one.
+
+2. Fixed again a compilation problem when using old e2fs
+ headers (filesystem label related). Thanks to many users
+ who reported this stupid error.
+
+3. Fixed a build problem on old lib5 based systems dealing with
+ _PATH_MNTTAB being undefined. Thanks to John Adams
+ <johna@onevista.com> for reporting the bug.
+
+4. Improved the error detection in restore when creating
+ the temporary files in TMPDIR. Now restore will corectly
+ report a 'no space left on device' error instead of
+ strange errors which could imply an archive corruption.
+ Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
+ bgjenero <bgjenero@sympatico.ca> for reporting the bug.
+
+5. Added the throughput information in dump's progression
+ messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
+ for the patch.
+
+6. Use libext2fs's inode scanning routines, which are particularly
+ robust when dealing with errors (especially when having some
+ bad blocks on the filesystem). Many thanks to Colin
+ <colin@horizon.com> for reporting the bug, investigating
+ the issues, finding a workaround, writing the patch and
+ fully testing it... (of course, if this patch will break
+ anything, he is to blame for :-)).
+
+7. Made dump and restore LFS aware. Dump can dump a filesystem
+ containing large files, generate a large file on output and
+ restore can restore them. This feature is not enabled by
+ default, use --enable-largefile option of configure to enable
+ it (you need to have a LFS aware glibc though). Thanks to
+ Andrea Arcangeli <andrea@suse.de> for submitting the patch,
+ and to Theodore T'so <tytso@valinux.com> for his always
+ useful thoughts.
+
+8. Made dump ask upon a tape write error if it should rewrite
+ the current volume (assume this is a bad tape) or if it should
+ assume an end-of-tape condition (useful for tape drives which
+ behaves badly at the end of the tape). Thanks to Andreas
+ Dilger <adilger@turbolinux.com> for the suggestion.
+
+Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
+=======================================================================
+
+1. Fixed a small compilation problem due to a change
+ in the definintion of the struct sigaction in
+ glibc 2.0 / libc5. Thanks to Gunther Schlegel
+ <schlegel@riege.de> for reporting the bug and to
+ Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
+ a fix.
+
+2. Modified the label and uuid analysis in order to be
+ self-contained (independant of kernel/libc headers). This
+ should fix the compile with older kernel/libc headers and
+ will preserve the functionality. Thanks to Bernhard Erdmann
+ <bernhard.erdmann@gmx.de> for reporting the bug.
+
+3. The 'exclude inode' option, if applied to a directory
+ inode, excludes now correctly all the leaves of this
+ directory. Thanks to John R. Dennison
+ <gerdesas@users.sourceforge.net> for reporting the bug.
+
+4. Fixed the '-e' option to disable the possibility
+ to exclude the root inode (which causes the dumps to
+ be unreadable by restore). Prevented array overflow
+ when multiple -e options are used.
+
+5. Fixed dump to correctly interpret a filesystem argument
+ which is a mountpoint and it is not an absolute path
+ (as specified in the fstab). Thanks to Bernhard R. Erdmann
+ <be@berdmann.de> for reporting the bug.
+
+6. Made dump able to backup files larger than 2 GB. Note that
+ dump still doesn't cope with files larger than 4 GB.
+
+7. Restore the real uid and gid before invoking an external
+ RSH program (big hole when dump or restore is suid root!).
+
+8. Get the values for the default tape device and the location
+ of fstab file from the system headers. Thanks to
+ Andreas Dilger <adilger@turbolinux.com> for the patch.
+
+9. Made dump -w|-W report all recognized filesystems
+ present in either /etc/fstab or /etc/dumpdates, and present
+ the list in the same order as in fstab file. Thanks
+ to Andreas Dilger <adilger@turbolinux.com> for the patch.
+
+10. Made dump's -a (automatic end of tape detection) the
+ default. Specifying one of -B, -c, -d or -s options will
+ override the EOM detection. Thanks to Andreas Dilger
+ <adilger@turbolinux.com> for the patch.
+
+11. Save the ext2 filesystem label into the dump volume label.
+ Specifying a volume label on the command line (-L option)
+ will override this feature. Thanks to Andreas Dilger
+ <adilger@turbolinux.com> for the patch.
+
+Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
+=====================================================================
+
+1. Fixed the signal handling in dump (which I broke in 0.4b17)
+ which was causing several strange dump failures (dump
+ hanged or segmentation faults etc).
+
+2. Specified the default blocksize in the dump man page.
+
+3. Changed two info messages of restore to be written on stdout
+ instead of stderr in order to leave stderr only for errors
+ or warnings. Thanks to Stephen Atwell
+ <satwell@urbana.css.mot.com> for the suggestion.
+
+4. Corrected an off by one calculation which prevented
+ dumping a directory having a 1 character name length.
+ Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
+ for reporting the bug.
+
+5. Reinforce dump in order to allow only level 0 and no
+ -u option when dumping a subdirectory, and document
+ this in the man page. Thanks to Bernhard Erdmann
+ <bernhard.erdmann@gmx.de> for reporting the bug.
+
+6. Fixed a small harmless bug in restore which caused
+ in some conditions hard links to be restored several
+ times (therefore generation some warning messages).
+ Thanks to Kenneth Porter <shiva@well.com> for
+ reporting the bug.
+
+7. Updated the RPM spec file to the latest RedHat version,
+ providing FHS packaging and other cosmetic changes.
+ You will need RPM version >= 3.0.5 in order to build the RPM.
+
+8. Updated the configure script to check for libtermcap
+ before checking for libreadline, because we need this
+ library in order to compile the readline support.
+
+9. Made dump understand the LABEL= and UUID= notation
+ both in /etc/fstab and on the command line. Note that
+ you will need the /proc filesystem in order to use
+ these notations. Thanks to Erik Troan <ewt@redhat.com>
+ for providing the patch.
+
+Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
+===================================================================
+
+1. Fixed a potential buffer overflow in restore. Thanks
+ to Stan Bubrouski <satan@fastdial.net> for reporting
+ the bug.
+
+2. Fixed a readline-related bug which prevented
+ 'cat DUMPFILE | restore -i -f -' from working. Thanks
+ to Charles Karney <karney@users.sourceforge.net>
+ for the bug report.
+
+3. Changed a few "panic" into "exit", causing restore to
+ be more stable against some attacks (like the last one
+ reported on Bugtraq, although the last version of restore
+ was not vulnerable - just dumped core). Thanks to
+ Andreas Hasenack <andreas@conectiva.com.br> for reporting
+ the bugs to me.
+
+4. Removed the suid-root bit on dump and restore in the
+ default build (and generated RPMs). It should be safer
+ now. Users who need the suid-root capabilities in order
+ to make network backups should read first the man page
+ of dump and enable the suid bit manually.
+
+5. Added -ltermcap to the compile parameters for restore
+ when using readline support, in order to make the compile
+ process work on some systems (SuSE ?). Thanks to
+ Patrik Schindler <poc@pocnet.net> for reporting the bug.
+
+Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
+====================================================================
+
+1. The -F script is called now *only* at the end of a tape,
+ not at the end of the dump. Thanks to Christian Haul
+ <haul@informatik.tu-darmstadt.de> for the bug report.
+
+ Normally, the device name and the current volume number
+ are passed on the command line of the script. However,
+ if you want the old style script (with no arguments
+ passed on the command line), you can enable it in
+ configure with the --enable-oldstylefscript.
+
+2. Use posix signal handling to preserve dump functionality
+ with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
+ the patch.
+
+3. Made the exit code of restore in case of a 'C'ompare
+ command reflect the result of the compare. An exit status
+ of 0 means the dump archive is correct, 1 means tape errors,
+ 2 means that some files were modified. Thanks to Kenneth Porter
+ <shiva@well.com> for the suggestion.
+
+4. Made (finally) quotes work in interactive restore.
+
+5. Small fixes in order to allow dump to compile with a
+ really old e2fsprogs version. Thanks to Ian Zimmerman
+ <itz@speakeasy.org> for the bug report.
+
+6. Add GNU readline capabilities to interactive restore.
+ Use configure's --enable-readline flag to enable this feature.
+ Thanks to Patrik Schindler <poc@pocnet.net> for the
+ suggestion.
+
+7. Do the compare on the fly in restore 'C' mode (this will
+ allow not to exhaust the available /tmp space when
+ ccmparing large files). Thanks to Kenneth Porter
+ <shiva@well.com> for the suggestion.
Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
====================================================================