hi, i'm making a program that reads from a rsync log file, extracts useful information and writes it to a html file. the problem is: i don't know the way to identify a rsync block execution, that is a block of text in the log that have the same id process module, and source directory. i can't use the process id alone, 'cause, although it identifies the block, it can be used again for other rsync excecution. i readed the documentation and i did'n find a way to give an id to the block (other than the process id). can anyone help me? i can't alter the rsync source code neither(company's politic) -------------- next part -------------- HTML attachment scrubbed and removed
I am attempting to set up a "backup server" using rsync - it will store the backups from my 5 other servers and initiate the rsync process to pull the information over to itself. From there, it will all be going to tape. I've read just about everything I could the past few weeks, and feel I am very close to a solution, but google's not helping much.>From the backup server, I initiate the rsync process with the command:rsync -av 10.10.10.5::dns/ /mnt/dns1 On the server being backed up, I have the following rsync.conf file: uid = 0 gid = 0 use chroot = no max connections = 4 syslog facility = local5 pid file = /var/run/rsyncd.pid read only = true list = false hosts allow = 65.19.234.235 [dns] path = /var/named comment = dns config files [web] path = /var/www comment = test web files My problem: I keep getting the following error and can't find much info on it, or how to fix it: rsync: read error: connection reset by peer rsync error: error in rsync protocol data stream (code 12) at io.c(177) I've started the rsync process with rsync --daemon, and have modified the appropriate services and xinetd files. I've also sent a sig hup to xinetd. Any help, or a point the in the right direction would be greatly appriciated. -------------- next part -------------- HTML attachment scrubbed and removed
On Tue, Sep 13, 2005 at 02:57:34PM -0300, Mario Tambos wrote:> i can't use the process id alone, 'cause, although it identifies the > block, it can be used again for other rsync excecution.Firstly, PIDs on most systems don't repeat very quickly, so just using a PID combined with the date should be unique enough for most purposes (though you'd need some extra logic to avoid splitting a session that crosses midnight). If that's not enough, you can note the distinct starting message that rsync always uses at the start of a session ("rsync on mod/path from host...") and create a unique ID based on the PID and the time of day when that occurs. You'd then map all other messages with that PID to your unique ID up until a starting message is seen again for that PID, which would create a new PID -> unique-ID mapping. ..wayne..