patch provided by Daniel Neri in PR 32777
New in v1.0.4 (2006/01/15)
--------------------------
Files with ACLs will not be unnecessarily marked as changed (bug
report by Carsten Lorenz).
Another fix for long-filenames crash, this time when a long-named
directory with files in it gets deleted
Selection fix: empty directories could sometimes be improperly
excluded if certain include expressions involving a non-trailing '**'
were used. Bug reported by Toni Price.
Bug #15436 reported by Remy Blank: rdiff-backup would sometimes crash
on ACLs if the owning uname did not exist on destination.
New in v1.0.3 (2005/11/25)
--------------------------
Applied Alec Berryman's patch to update the no-compression regexp.
Alec Berryman's fs_abilities patch is supposed to help with AFS.
Fixed (version of) filename-too-long crash when quoting.
Due to very detailed error report from Yoav, fixed a "Directory not
empty" error that can arise on emulated filesystems like NFS and
EncFS.
Cleaned up remove older than report, and also stopped it from deleting
current data files if you specify a time later than the current
mirror.
--compare fix, now you don't need to specify the exact time of a
session to avoid a "no metadata" error.
New in v1.0.2 (2005/10/24)
--------------------------
Fix for spurious security violation from --create-full-path (reported
by Mike Bydalek).
Fix for bug 14545 which was introduced in version 1.0.1: Quoting
caused a spurious security violation. (Important for Mac OS X)
An error reading carbonfile data on Mac OS X should no longer cause a
crash. (Thanks to Kevin Horton for testing.)
Carbonfile support now defaults to off, even if the system appears to
support it. It can be manually enabled with the --carbonfile switch.
If you know something about Mac OS X and want to look at the
carbonfile code so it can be re-enabled by default, please do so :)
(help available from list)
Fix for bug #14799 reported by Bob McKay: Crash when backing up files
with high permissions (like suid) to some FAT systems.
RECOMMENDED is removed. It becomes ABI_DEPENDS.
BUILDLINK_RECOMMENDED.foo becomes BUILDLINK_ABI_DEPENDS.foo.
BUILDLINK_DEPENDS.foo becomes BUILDLINK_API_DEPENDS.foo.
BUILDLINK_DEPENDS does not change.
IGNORE_RECOMMENDED (which defaulted to "no") becomes USE_ABI_DEPENDS
which defaults to "yes".
Added to obsolete.mk checking for IGNORE_RECOMMENDED.
I did not manually go through and fix any aesthetic tab/spacing issues.
I have tested the above patch on DragonFly building and packaging
subversion and pkglint and their many dependencies.
I have also tested USE_ABI_DEPENDS=no on my NetBSD workstation (where I
have used IGNORE_RECOMMENDED for a long time). I have been an active user
of IGNORE_RECOMMENDED since it was available.
As suggested, I removed the documentation sentences suggesting bumping for
"security" issues.
As discussed on tech-pkg.
I will commit to revbump, pkglint, pkg_install, createbuildlink separately.
Note that if you use wip, it will fail! I will commit to pkgsrc-wip
later (within day).
developer is officially maintaining the package.
The rationale for changing this from "tech-pkg" to "pkgsrc-users" is
that it implies that any user can try to maintain the package (by
submitting patches to the mailing list). Since the folks most likely
to care about the package are the folks that want to use it or are
already using it, this would leverage the energy of users who aren't
developers.
Lots of changes/bugfixes including:
rdiff-backup now writes its PID to current_mirror marker
Security fixes with --restrict-read-only, --restrict-update-only and
--restrict
OSX fixes
For a full list of changes see:
http://www.nongnu.org/rdiff-backup/CHANGELOG-stable
python*-pth packages into meta-packages which will install the non-pth
packages. Bump PKGREVISIONs on the non-pth versions to propagate the
thread change, but leave the *-pth versions untouched to not affect
existing installations.
Sync all PYTHON_VERSIONS_AFFECTED lines in package Makefiles.
in the process. (More information on tech-pkg.)
Bump PKGREVISION and BUILDLINK_DEPENDS of all packages using libtool and
installing .la files.
Bump PKGREVISION (only) of all packages depending directly on the above
via a buildlink3 include.
Ok'ed wiz@/snj@/davids (at) idiom.com
From the CHANGELOG:
New in v0.12.7 (??????????)
---------------------------
Altered file selection when restoring so excluded files will not be
deleted from the target dir. The old behavior was technically
intended and documented but not very convenient. Thanks to Oliver
Kaltenecker for bug report.
Fixed error when --restrict path given with trailing backslash. Bug
report by Åke Brännström.
Backported regress fix from 0.13.x. Bug noticed by Alan Horn.
Added error-correcting fsync suggestion by Antoine Perdaens.
rdiff-backup may work better with NFS now.
Fix for regress warning code: rdiff-backup should warn you if you are
trying to back up a directory into itself.
Changes since 0.12.5:
* Yet another unreadable file non-root regress bug fix.
* Added --list-increment-sizes option from the development branch.
Previously this option was in the man page but was omitted in the code
(thanks to Michael Schwendt for report).
* Regressing and restoring should now take less memory when processing
large directories (noticed by Luke Mewburn and others).
* When regressing, remove mirror_metadata and similar increments first.
This will hopefully help regressing a backup that failed because disk
was full (reported by Erik Forsberg).
* Fixed rather important quoting problem: quoting directives like
--windows-mode were simply ignored when rdiff-backup was running
remotely! I'm surprised no one noticed this. Are none of you using
--windows-mode or similar?
* Fixed handling of (lack of) daylight savings time. Earlier bug would
cause some files to be marked an hour later. Thanks to Troels Arvin
and Farkas Levente for bug report.
New in v0.12.5 (2003/09/27)
---------------------------
Fixed bug in --test-server when using --restrict security options
(test would fail improperly). Thanks to Maik Schreiber for report.
--list-changed-since and --list-at-time now work remotely.
Thanks to Morten Werner Olsen for bug report.
Fixed logic bug that could make restoring extremely slow and waste
memory. Thanks for Jacques Botha for report.
Fixed bug restoring some directories when mirror_metadata file was
missing (as when made by 0.10.x version).
Regressing and restoring as non-root user now works on directories
that contain unreadable files and directories as long as they are
owned by that user. Bug report by Arkadiusz Miskiewicz. Hopefully
this is the last of the unreadable file bugs...
New in v0.12.4 (2003/09/13)
---------------------------
Specified socket type as SOCK_STREAM. (Error reported by Erik
Forsberg.)
Fixed bug backing up unreadable regular files and directories when
rdiff-backup is run by root on the source site and non-root on the
destination side. (Reported by Troels Arvin and Arkadiusz
Miskiewicz.)
If there is data missing from the destination dir (for instance if a
user mistakenly deletes it), only warn when restoring, instead of
exiting with error.
Fixed problems with --restrict options that would cause proper
sessions to fail. Thanks to Randall Nortman for error report.
File examples.html added to distribution; examples section removed
from man page.
Inspired by FreeBSD "ports".
Fix the PLISTs accordingly.
Also, while at it, remove now obsolete compileall.py calls in post-install
targets and insure that extension.mk is in included before builinlinks of
other Python modules.
Discussed with/ok'ed by drochner@.
- Over 1500 lines of changelog since 0.10.2
- Now rdiff-backup writes metadata (uid, gid, mtime, etc.) to a
compressed text file in the rdiff-backup-data directory
- No longer seems compelled to send symlinks every time
Provided in PR 18577 by David.S at idiom dot com, some modifications
by me to use buildlink2 files, and to specify the correct version of
python required.
Rdiff-backup backs up one directory to another, possibly over a network.
The target directory ends up a copy of the source directory, but extra
reverse diffs are stored in a special subdirectory of that target directory,
so you can still recover files lost some time ago. The idea is to combine
the best features of a mirror and an incremental backup. Rdiff-backup also
preserves subdirectories, hard links, dev files, permissions, uid/gid
ownership (if it is running as root), and modification times. Finally,
rdiff-backup can operate in a bandwidth efficient manner over a pipe, like
rsync. Thus you can use rdiff-backup and ssh to securely back a hard drive
up to a remote location, and only the differences will be transmitted.
