Re: Revised flags patch

2008-02-15 Thread Rolf Grossmann
Wayne Davison wrote: Your patch calls chflags(file,0) on every file/dir that is going to be removed or renamed. Does that slow things down at all? I didn't do any benchmark, but it doesn't appear that bad to me. It's one extra syscall per unlink/rename, but I guess the time spent on i/o is

Revised flags patch

2008-02-15 Thread Rolf Grossmann
Hi, first of all, sorry for taking so long. Unfortunately, some other tasks kept coming up. Anyway, attached is the version of the flags patch, that is based on the one I'm using with 2.6.9. It is against the rsync-3.0.0pre9 release. I've included the option name change from the repository, so th

Small oversight in hard_link_one

2008-02-06 Thread Rolf Grossmann
Hi, while looking over my flags patch for the upcoming 3.0 version (more on that later), I noticed a small bug was introduced into hard_link_one when changing the return codes. As a result, in the rare case of the hard link failing and the verbose option not in use, rsync wouldn't try to copy the

Re: 4.4BSD chflags support for rsync

2002-04-25 Thread Rolf Grossmann
Hi Dave, on Thu, 25 Apr 2002 15:52:56 -0500 Dave Dykstra wrote concerning "Re: 4.4BSD chflags support for rsync" something like this: > I think the --flags patch is probably fine, except that that the short > description should at least say "preserve file flags" instead of just > "preserve flag

Re: 4.4BSD chflags support for rsync

2002-03-08 Thread Rolf Grossmann
Hi Dave, on Fri, 8 Mar 2002 11:55:10 -0600 Dave Dykstra wrote concerning "Re: 4.4BSD chflags support for rsync" something like this: > New options are usually handled by just letting the remote side fail when > it doesn't understand the option, as you're doing. Perhaps you just > shouldn't inc

Re: --exclude not used in all cases?

2000-11-21 Thread Rolf Grossmann
Hi, on Tue, 21 Nov 2000 15:11:22 -0600 Dave Dykstra wrote concerning "Re: --exclude not used in all cases?" something like this: > Looks like the include "/*" from the server side is overriding the > exclude "crap" on the client side. If it were in a subdirectory it > would probably be exclude

--exclude not used in all cases?

2000-11-21 Thread Rolf Grossmann
Hi, I'm having a problem with the --exclude option of rsync. It looks to me like it is ignored on the client side when transfering files from the server (that is only for files that exist on the server). Let me show you my example setup: /etc/rsyncd.conf: [test] comment = Test path = /tmp/rsyn