[softwarelist] Re: SparkFS 1.41 & RO 6

  • From: Dave Symes <dfs@xxxxxxxxxxxxx>
  • To: davidpilling@xxxxxxxxxxxxx
  • Date: Fri, 01 Jun 2007 07:15:35 +0100

In article <1d04d5eb4e.martin@xxxxxxxxxxxxxxxxxxx>,
   Martin Wuerthner <public@xxxxxxxxxxxxxxx> wrote:
> In message <4eebc7a113dfs@xxxxxxxxxxxxx>

[Snip]

> > I still don't understand how a version of SparkFS that is not named
> > SparkFS could steal the path pointer from a version set in Boot
> > Configure Look at, as SparkFS.

> > Ah well such are the mysteries of computing.

> Not a mystery at all. The name of the application does not mean 
> anything. If your other copy is "seen" by the Filer after the one 
> called SparkFS, it takes precedence. That is standard RISC OS 
> behaviour. Presumably, the reason why a copy of SparkFS is there under 
> a different name is because you have renamed it. You should either 
> refrain from doing so or you should know what the consequences are. 
> The bottom line is to never allow outdated applications to hang around 
> where they are seen by the Filer. The best way to ensure that is to 
> keep them as a standard directory without a leading "!". For instance, 
> supposing you have SparkFS 1.30 and you want to keep a backup when you 
> upgrade to SparkFS 1.41. Then, simply take a copy as SparkFS130 (NB - 
> without a leading "!") and upgrade the original !SparkFS directory to 
> the newer version.

> Martin

Thanks for the explanation Guys, now you've put the words down, so to
speak, I can see what happened.

After the RO 6 install, I moved a copy of the other !SparkFSNC out of its
lower level directory, up one level to the same dir in which resides the
in use !SparkFS and forgot I'd done that action, consequently because of
its name it was seen after the SparkFS...

Now that it is back down it's okay.

Backup on this machine are not a problem, as they are stored on the second
Hardrive which is only used for backing up.

Martin, I can't change it, it is the same version as the top level
SparkFS, it's just configured very differently, so when needed I can just
quit the standard SparkFS and run the differently configured version.

I have instead now removed the pling from the !Boot file inside the
SparkFSNC version.

Again, thanks guys for the help.

Cheers
Dave S

-- 

Other related posts: