X-Git-Url: https://mattmccutchen.net/rsync/rsync.git/blobdiff_plain/42e9c7eb565a6084d5a11c1808f5837babd7955b..99eba67585c79180be72ea10889ae7eec61f9034:/NEWS diff --git a/NEWS b/NEWS index 7e60adde..50e12ac1 100644 --- a/NEWS +++ b/NEWS @@ -1,85 +1,109 @@ -NEWS for rsync 2.6.5 (UNRELEASED) +NEWS for rsync 2.6.9 (unreleased) Protocol: 29 (unchanged) -Changes since 2.6.4: +Changes since 2.6.8: BUG FIXES: - - 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. + - If rsync is interrupted via a handled signal (such as SIGINT), it will + once again clean-up its temp file from the destination dir. - - Fixed a bug in the transfer of a single file when -H is specified - that might have caused an infininte loop or perhaps a crash. + - Fixed an overzealous sanitizing bug in the handling of the --link-dest, + --copy-dest, and --compare-dest options to a daemon without chroot: if + the copy's destination dir is deeper than the top of the module's path, + these options now accept a safe number of parent-dir (../) references + (since these options are relative to the destination dir). The old code + incorrectly chopped off all "../" prefixes for these options, no matter + how deep the destination directory was in the module's hierarchy. - - Fixed a case where the generator might try to tweak the write - permissions of a read-only directory in list-only mode. + - Fixed a bug where a deferred info/error/log message could get sent + directly to the sender instead of being handled by rwrite() in the + generator. This fixes an "unexpected tag 3" fatal error, and should + also fix a potential problem where a deferred info/error message from + the receiver might bypass the log file and get sent only to the client + process. (These problems could only affect an rsync daemon that was + receiving files.) - - 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. + - Fixed a bug where using --dry-run with a --*-dest option with a path + relative to a directory that does not yet exist: the affected option + gets its proper path value so that the output of the dry-run is right. - - Fixed the accidental disabling of --backup during the --delete-after - processing. + - Fixed a bug in the %f logfile escape when receiving files: the + destination path is now included in the output (e.g. you can now tell + when a user specifies a subdir inside a module). - - Restored the ability to use the --address option in client mode (in - addition to its use in daemon mode). + - If the receiving side fails to create a directory, it will now skip + trying to update everything that is inside that directory. - - 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. + - If --link-dest is specified with --checksum but without --times, rsync + will now allow a hard-link to be created to a matching link-dest file + even when the file's modify-time doesn't match the server's file. - - When --existing skips a directory with extra verbosity, don't refer - to it as a "file". + - The daemon now calls more timezone-using functions prior to doing a + chroot. This should help some C libraries to generate proper timestamps + from inside a chrooted daemon (and to not try to access /etc/timezone + over and over again). - - When transferring a single file to a different-named file, any - generator messages that are source-file related no longer refer to - the file by the destination filename. + - Fixed a bug in the handling of an absolute --partial-dir=ABS_PATH option: + it now deletes an alternate basis file from the partial-dir that was used + to successfully update a destination file. - - Fixed a bug where hard-linking a group of files might fail if the - generator hasn't created the destination directory yet. + - Fixed a bug in the handling of --delete-excluded when using a per-dir + merge file: the merge file is now honored on the receiving side, and + only its unqualified include/exclude commands are ignored (just as is + done for global include/excludes). - - Fixed a bug where a hard-linked group of files that is newly-linked - to a file in a --link-dest dir doesn't link the files from the rest - of the cluster. + - Fixed a recent bug where --delete was not working when transferring from + the root (/) of the filesystem with --relative enabled. - ENHANCEMENTS: + - Fixed a recent bug where an --exclude='*' could affect the root (/) of + the filesystem with --relative enabled. + + - When --inplace creates a file, it is now created with owner read/write + permissions (0600) instead of no permissions at all. This avoids a + problem continuing a transfer that was interrupted (since --inplace + will not update a file that has no write permissions). - - 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). + - If either --remove-source-files or --remove-sent-files is enabled and we + are unable to remove the source file, rsync now outputs an error. - - 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 "\\". + - Fixed a bug in the daemon's "incoming chmod" rule: newly-created + directories no longer get the 'F' (file) rules applied to them. - - 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.) + ENHANCEMENTS: - - 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). + - Added the --log-file=FILE and --log-file-format=FORMAT options. These + can be used to tell any rsync to output what it is doing to a log file. + They work with a client rsync, a non-daemon server rsync (see the man + page for instructions), and also allows the overriding of rsyncd.conf + settings when starting a daemon. - - 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). + - The --log-format option was renamed to be --out-format to avoid confusing + it with affecting the log-file output. (The old option remains as an + alias for the new to preserve backward compatibility.) - INTERNAL: + - Made "log file" and "syslog facility" settable on a per-module basis in + the daemon's config file. - - Rsync now calls setlocale(LC_CTYPE, ""). + - Added the --remove-source-files option as a replacement for the (now + deprecated) --remove-sent-files option. This new option removes all + non-dirs from the source directories, even if the file was already + up-to-date. This fixes a problem where interrupting an rsync that + was using --remove-sent-files and restarting it could leave behind + a file that the earlier rsync synchronized, but didn't get to remove. + (The deprecated --remove-sent-files is still understood for now, and + still behaves in the same way as before.) - BUILD CHANGES: + - Added a new environment variable to the pre-/post-xfer exec commands (in + the daemon's config file): RSYNC_PID. This value will be the same in + both the pre- and post-xfer commands, so it can be used if the pre-xfer + command wants to cache some arg/request info for the post-xfer command. - - Fixed a bug in the SUPPORT{,_HARD}_LINKS defines which prevented - rsync from being built without symlink or hard-link support. + DEVELOPER RELATED: - - You can use --disable-locale to turn off any use of setlocale(). + - The acls.diff and xattrs.diff patches have received a bunch of work to + make them much closer to being acceptable in the main distribution. - - Configure now disables the use of mkstemp() under HP-UX (since they - refuse to fix its broken handling of large files). + - Updated config.guess and config.sub to their 2006-02-23 version. - - 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. + - Updated various files to include the latest FSF address and to have + consistent opening comments.