RFE blues (Re: [Bug 8188] Mechanism for taking an rsync server down for maintenance)

Matt McCutchen matt at mattmccutchen.net
Thu Jun 2 09:23:42 MDT 2011

On Wed, 2011-06-01 at 17:08 -0400, Brian K. White wrote:
> On 6/1/2011 3:26 PM, Matt McCutchen wrote:
> > On Wed, 2011-06-01 at 14:57 -0400, Brian K. White wrote:
> >> I like the built-in idea as I don't happen to use rsync via inetd/xinetd
> >> or any other on-demand starter.
> >>
> >> It's not an actual problem for me, today, but that's no excuse to avoid
> >> doing the right thing once you recognize it.
> >
> > And "the right thing" for upstream rsync is in Wayne's sole discretion.
> > In my view, the perl script is a fine solution and what convenience
> > there may be in a built-in option does not merit the effort of
> > maintaining the feature as part of rsync.
> >
> That is a work-around, not a solution.

The difference between the two is a matter of opinion.  I don't share

> The work-around requires using (x)inetd

This is not true of the program in comment 1.

> If the service aims to be usable stand-alone, which rsync clearly does, 
> then imo this is just part of that. A nicity perhaps, not critical, but 
> trivial to implement and maintain in comparison with things like 
> child/session management which are definitely in there solely to support 
> standalone daemon operation.

Right.  Feel free to maintain an enhanced version of rsync with every
proposed feature that is straightforward to maintain in isolation and
potentially helpful to rsync's mission, and if you do a good enough job
the distros will surely adopt it.  Wayne will maintain what he judges is
worthwhile; you should not expect him to do otherwise, unless you hire


More information about the rsync mailing list