libfor consistency use memcpy/memset everywhere instead of bcopy/bzero 28 years ago packagingpreparing for release of 2.1.0 28 years ago zlibreplace BAD with zBAD so it compiles on AIX 28 years ago .cvsignoreadded .cvsignore to the .cvsignore file to prevent it from being distributed when I run the release script 28 years ago .ignoreInitial revision 30 years ago access.cadd a cast to initialisation of mask 28 years ago acconfig.hhandle hard links on systems with 16 bit ino_t 28 years ago authenticate.cfor authenticated access record the authenticated username in the logs 28 years ago byteorder.hInitial revision 30 years ago checksum.cfor consistency use memcpy/memset everywhere instead of bcopy/bzero 28 years ago clientserver.c- defer the error message from the options parsing until after the socket is multiplexed. This allows clients sending new options which the remote server doesn't understand to get a sensible error message. 28 years ago compat.cadded a bit in the man page about the clean shell error 28 years ago configure.inhandle hard links on systems with 16 bit ino_t 28 years ago connection.cuse error to detect lockfile open failures vs. max connections reached and report an appropriate error message 28 years ago COPYINGInitial revision 30 years ago exclude.cuse Realloc instead of realloc 28 years ago fileio.cheaps of cleanup in the io code. 28 years ago flist.cmake rsync behave more like GNU cp with regard to file permissions when -p (preserve permissions) isn't set. 28 years ago hlink.cfor consistency use memcpy/memset everywhere instead of bcopy/bzero 28 years ago install-shInitial revision 30 years ago io.cI think I might havefinally fixed the rsync hanging bug. It was caused by a read during an io_flush() triggered during a readfd(). A simple logic bug in the io code :( 28 years ago loadparm.cfor consistency use memcpy/memset everywhere instead of bcopy/bzero 28 years ago log.c- use explicit flushes instead of setlinebuf. I've had reports of verbose info not being line buffered to files. 28 years ago main.c- close stdout and stderr and reopen then as /dev/null when running as a daemon. This prevents library functions (such as getopt) stuffing up our protocol stream when errors are detected. 28 years ago Makefile.inremoved old "make dist" target 28 years ago match.c- don't show "created directory" message unless verbose is selected 28 years ago md4.csome large systematic changes to make socket conversion easier. The biggest one is the use of rprintf() instead of fprintf() 28 years ago md4.hInitial revision 30 years ago mkproto.awkfirst vesrion of working socket based rsync. It still needs a lot of work, but at least it works :-) 28 years ago options.c- close stdout and stderr and reopen then as /dev/null when running as a daemon. This prevents library functions (such as getopt) stuffing up our protocol stream when errors are detected. 28 years ago params.cuse Realloc instead of realloc 28 years ago READMEupdated the README a bit 28 years ago rsync.cI think I might havefinally fixed the rsync hanging bug. It was caused by a read during an io_flush() triggered during a readfd(). A simple logic bug in the io code :( 28 years ago rsync.hhandle hard links on systems with 16 bit ino_t 28 years ago rsync.yoadded a bit in the man page about the clean shell error 28 years ago rsyncd.conf.yomanpage updates, mostly suggested by Francois 28 years ago socket.cI think I might havefinally fixed the rsync hanging bug. It was caused by a read during an io_flush() triggered during a readfd(). A simple logic bug in the io code :( 28 years ago syscall.ccheck for munmap as well as mmap. NextStep only has mmap in standard libs 28 years ago tech_report.texInitial revision 30 years ago test.shadded some notes to test.sh 28 years ago token.cUpdate to use the new zlib-1.1.2 code. The compressed token code now handles the null (-2) token from the match logic. 28 years ago uidlist.cchange to allow names or numbers to be used for uid and gid. 28 years ago util.cI think I might havefinally fixed the rsync hanging bug. It was caused by a read during an io_flush() triggered during a readfd(). A simple logic bug in the io code :( 28 years ago version.hpreparing for release of 2.1.0 28 years ago