[sharp-discuss] installation error "during host resources" on Mac OS X 10.5.2

Peter Keller pkeller at globalphasing.com
Thu May 8 12:10:19 BST 2008


Dear Eugene,

On Thu, 8 May 2008, Eugene Valkov wrote:

> Hi everyone,
>
> I'm currently battling with a SHARP/autoSHARP server install on Mac OS X 
> 10.5.2 using the downloaded version of the program for Darwin/MacOsX. 
> Basically, everything goes smoothly with a fast install (./installSHARP -F) 
> from a non-root account (although the same happens during root installation) 
> until the section 6.1 "server setup" of the installation when the process 
> aborts with the following message:
>
>
> NOTE : We will now generate a basic server setup file. Make
>      sure to answer the questions accurately - otherwise
>      the interface SUSHI may not work properly.
>
>
> NOTE : You can re-configure your server by running
>      "host-setup".
>
>
> NOTE : The SHARP / SUSHI server will be set to the machine
>      eugene
>
> -n         Continue ? [y]es/[n]o/[e]xit/[s]kip >
> y
>
> ERROR : unable to use the "echo -n" wrapper

I don't know the details of how this particular script works, but it looks as 
if this arises from Apple having changed Leopard to conform to the Single Unix 
Specification. See for example:

   http://www.macosxhints.com/article.php?story=20071106192548833

There are a couple of likely-looking approaches listed in this post that might 
be worth a try (e.g. editing the script to use /bin/echo instead of echo). 
Unfortunately none of us have got our hands on Leopard, so we can't try it out 
ourselves.

> ERROR : during host resources
>
>
>
> Can anybody shed light on this? Or am I better off with linux? ;-)

If your mac is an intel one and you are not short of RAM, you could try using 
a linux virtual machine. I have tried this using VMWare Fusion in Tiger, and 
SHARP runs a _lot_ faster this way than using our PPC binary (more than twice 
as fast). The only caveat is that I/O from a VMWare machine using NFS or the 
VMWare shared folder mechanism is relatively slow. This isn't limiting for 
just running SHARP, but it may be a consideration if you are moving a lot of 
data around.

Good luck,
Peter.



More information about the sharp-discuss mailing list