Subject: Re: pkg/15501: rsync development branch causes corruption with alpha server
To: Tracy J. Di Marco White <gendalia@iastate.edu>
From: Greg A. Woods <woods@weird.com>
List: netbsd-bugs
Date: 02/05/2002 21:54:44
[ On Tuesday, February 5, 2002 at 18:46:32 (CST), Tracy J. Di Marco White wrote: ]
> Subject: Re: pkg/15501: rsync development branch causes corruption with alpha server 
>
> woods@weird.com (Greg A. Woods) wrote:
> } [ On , February 5, 2002 at 21:24:26 (-0000), gendalia@iastate.edu wrote: ]
> } > Subject: pkg/15501: rsync development branch causes corruption with alpha server
> } >
> } > 2.5.x is the development branch of rsync.  2.4.8 is the patched stable
> } > branch version of rsync.
> }
> } This is not correct.
> 
> See http://freshmeat.net/projects/rsync/ "Branches".  They say it is.

You are confusing the weird and wonderful FreshMeat concept of
"branches" with the real-live actual branches in the official CVS
repository, and the associated documentation and commit messages in that
repository and the files it contains, and the status of the actual
releases as should be clearly seen in the download directory names.

Many people seem to use the "development" tag on FreshMeat to indicate
where development will actually procede from (as opposed to very minor
or necessary maintenance on the so-called "stable" branch), and given
the wealth of other evidence in this case that must be the meaning for
rsync.  "Stable" in this case probably really means "Stagnant".

Note also that on FreshMeat a "-current" kind of "branch" would be
marked with a name like "beta" or even "alpha".  The naming of branches
on FreshMeat is still somewhat of a dynamic thing, changing from the old
scheme (which is why many packages only have a "Default" branch, and
perhaps why the rsync "Development" branch contains many older "stable"
releases too).

> As far as I can tell, there are large problems with every version of
> rsync.

Perhaps -- though so far in 2.5.2. I've only encountered one
pseudo-problem with the new way exit codes are managed....

> I accept that it works better for you, it demonstrably does not work
> for the rsync server and mirrors for the project.  The most visible
> problems seem to be with LP64.

Did anyone try 2.5.0 on an LP64 machine and feed any bug reports back to
the maintainers?

-- 
								Greg A. Woods

+1 416 218-0098;  <gwoods@acm.org>;  <g.a.woods@ieee.org>;  <woods@robohack.ca>
Planix, Inc. <woods@planix.com>; VE3TCP; Secrets of the Weird <woods@weird.com>