-rsync 2.5.0 (2001-11-30) -*- indented-text -*-
+NEWS for rsync 2.6.5 (UNRELEASED)
+Protocol: 29 (unchanged)
+Changes since 2.6.4:
- ANNOUNCEMENTS
+ BUG FIXES:
- * Martin Pool <mbp@samba.org> is now a co-maintainer.
+ - A crash bug was fixed when a daemon had its "path" set to "/", did
+ not have chroot enabled, and used some anchored excludes in the
+ rsyncd.conf file.
- NEW FEATURES
+ - Fixed a case where the generator might try to tweak the write
+ permissions of a read-only directory in list-only mode.
- * Support for LSB-compliant packaging <http://www.linuxbase.org/>
+ - If --compare-dest or --link-dest uses a locally-copied file as the
+ basis for an updated version, log this better when --verbose or -i
+ is in effect.
- * Shell wildcards are allowed in "auth users" lines.
+ - Fixed the accidental disabling of --backup during the --delete-after
+ processing.
- * Merged UNC rsync+ patch to support creation of standalone patch
- sets. By Bert J. Dempsey and Debra Weiss, updated by Jos
- Backus. <http://www.ils.unc.edu/i2dsi/unc_rsync+.html>
+ - Restored the ability to use the --address option in client mode (in
+ addition to its use in daemon mode).
- * IPv6 support based on a patch from KAME.net, on systems
- including modern versions of Linux, Solaris, and HP-UX. Also
- includes IPv6 compatibility functions for old OSs by the
- Internet Software Consortium, Paul Vixie, the OpenSSH
- portability project, and OpenBSD.
+ - Make sure that some temporary progress information from the delete
+ processing does not get left on the screen when it is followed by a
+ newline.
- ENHANCEMENTS
+ - When --existing skips a directory with extra verbosity, don't refer
+ to it as a "file".
- * Include/exclude cluestick: with -vv, print out whether files are
- included or excluded and why.
+ ENHANCEMENTS:
- * Many error messages have more friendly explanations and more
- details.
+ - Added the --only-write-batch=FILE option that may be used (instead
+ of --write-batch=FILE) to create a batch file without doing any
+ actual updating of the destination. This allows you to divert all
+ the file-updating data away from a slow data link (as long as you
+ are pushing the data to the remote server when creating the batch).
- * Manual page improvements plus scanty protocol documentation.
+ - Changed the outputting of "safe" filenames to use backslash-escaped
+ characters rather than '?'s. Any non-printable character is output
+ using octal (e.g. "\n" -> "\012"), and backslash is output as "\\".
- * When running as --daemon in the background and using a "log
- file" rsyncd.conf directive, close the log file every time it is
- open when going to sleep on the socket. This allows the log
- file to get cleaned out by another process.
+ - Improved the auth-errors that are logged by the daemon to include
+ some information on why the authorization failed (wrong user,
+ password mismatch, etc.). (The client-visible message is unchanged.)
- * Change to using libpopt rather than getopt for processing
- options. This makes the code cleaner and the behaviour more
- consistent across platforms. popt is included and built if not
- installed on the platform.
+ - Improved the client's handling of an "@ERROR" from a daemon so that
+ it does not complain about an unexpectedly closed socket (since we
+ really did expect the socket to close).
- * More details in --version, including note about whether 64-bit
- files, symlinks and hardlinks are supported.
+ - If the daemon can't open the log-file specified in rsyncd.conf, fall
+ back to using syslog. This is better than a (typically) totally
+ silent failure (since a daemon is not usually run with --no-detach).
- * MD4 code may use less CPU cycles.
+ INTERNAL:
- * Use mkstemp on systems where it is secure. If we use mktemp,
- explain that we do it in a secure way.
+ - Rsync now calls setlocale(LC_CTYPE, "").
- * --whole-file is the default when source and target are on the
- local machine.
+ BUILD CHANGES:
- BUG FIXES:
+ - Fixed a bug in the SUPPORT{,_HARD}_LINKS defines which prevented
+ rsync from being built without symlink or hard-link support.
+
+ - You can use --disable-locale to turn off any use of setlocale().
+
+ - Configure now disables the use of mkstemp() under HP-UX (since they
+ refuse to fix its broken handling of large files).
- * Fix for various bugs causing rsync to hang.
-
- * Attempt to fix Large File Summit support on AIX.
-
- * Attempt to fix error handling lockup bug.
-
- * Give a non-0 exit code if *any* of the files we have been asked
- to transfer fail to transfer
-
- * For log messages containing ridiculously long strings that might
- overflow a buffer rsync no longer aborts, but rather prints an
- ellipsis at the end of the string. (Patch from Ed Santiago.)
-
- PLATFORMS:
-
- * Improved support for UNICOS (tested on Cray T3E and Cray SV1)
-
- * autoconf2.52 (or later) is now required to rebuild the autoconf
- scripts. It is not required to simply build rsync.
-
- * Platforms thought to work in this release:
-
- Cray SV1 UNICOS 10.0.0.8 cc
- Debian Linux 2.2 UltraSparc gcc
- Debian Linux testing/unstable ARM gcc
- FreeBSD 3.3-RELEASE i386 cc
- FreeBSD 4.1.1-RELEASE i386 cc
- FreeBSD 4.3-STABLE i386 cc
- HP PA-RISC HP-UX 10.20 gcc
- HP PA-RISC HP-UX 11.11 cc
- IRIX 6.5 MIPS cc
- IRIX 6.5 MIPS gcc
- Mac OS X PPC (--disable-ipv6) cc
- NetBSD 1.5 i386 gcc
- NetBSD Current i386 cc
- OpenBSD 2.5 Sparc gcc
- OpenBSD 2.9 i386 cc
- OpenBSD Current i386 cc
- RedHat 6.2 i386 gcc
- RedHat 6.2 i386 insure++
- RedHat 7.0 i386 gcc
- RedHat 7.1 i386 (Kernel 2.4.10) gcc
- Slackware 8.0 i686 (Kernel 2.4.10)
- Solaris 8 UltraSparc cc
- Solaris 8 UltraSparc gcc
- Solaris 8 i386 gcc
- SuSE 7.1 i386 gcc2.95.2
- SuSE 7.1 ppc gcc2.95.2
- i386-pc-sco3.2v5.0.5 cc
- i386-pc-sco3.2v5.0.5 gcc
- powerpc-ibm-aix4.3.3.0 cc
- i686-unknown-sysv5UnixWare7.1.0 gcc
- i686-unknown-sysv5UnixWare7.1.0 cc
-
- TESTING:
-
- * The existing test.sh script by Phil Hands has been merged into a
- test framework that works from both "make check" and the Samba
- build farm.
+ - Configure now explicitly checks for the lseek64() function so that
+ the code can use HAVE_LSEEK64 instead of inferring lseek64()'s
+ presence based on the presence of the off64_t type.