On Jul 21, Doug Williams wrote:
> Eli,
>
> I get a different error on my Linux box when I try to load the
> science collection:
>
> with-output-to-file: cannot open output file:
> "/usr/plt/collects/scribblings/main/user/.setup-plt-marker"
> (Permission denied; errno=13)
Ah, that's very useful.
On Jul 21, Doug Williams wrote:
> Eli, this seems to be related to the old Vista problem we had before
> (I think it was Robby that tracked it down) with the virtual store.
> On Vista, if system files are re-written, the updated copy is
> written in the Virtual Store - not to the file itself.
> Unfortunately, these persist over installs of PLT Scheme.
Yes, I remember that (I think that I was the one who talked about it
with you, or maybe it was someone else that I talked to), but this is
only partially related to the problem. The thing is that running a
planet setup uses the usual setup code, which is trying to verify that
all the collections to be setup are writeable by placing these marker
files -- and also using them to make sure that you're not trying to
setup the same collection twice (eg, `setup-plt -l foo foo').
One of the recent changes is that setting up a planet package is
properly reconstructing the user index, and this is somehow done in a
way that makes it (try to) lock the main/user collection.
Matthew -- is there some place in the code I can look at to see if
there's an easy solution?
--
((lambda (x) (x x)) (lambda (x) (x x))) Eli Barzilay:
http://barzilay.org/ Maze is Life!
_________________________________________________
For list-related administrative tasks:
http://list.cs.brown.edu/mailman/listinfo/plt-dev
