[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: More WSI stuff
- Subject: Re: More WSI stuff
- Date: Tue, 27 Mar 2001 10:41:13 -0700 (MST)
On Tue, 27 Mar 2001, Anne Wilson wrote:
> Robb Kambic wrote:
> >
> > On Mon, 26 Mar 2001, Anne Wilson wrote:
> >
> > > Hi Robb,
> > >
> > > On shemp, there is a 'request' line for WSI data - it looks more like
> > > the usual LDM configuration rather than the 'accept' and 'allow' setup
> > > that WSI said they recommend to their customers. I was going to try to
> >
> > Anne,
> >
> > Don't get confused, the wsihcsn machine is the WSI ingest machine. shemp
> > is a downstream machine so it uses a the standard request configurations.
> >
> > > set up imogene like Andrew Farnsworth had his machine set up (with the
> > > 'accept' and 'allow'),
> >
> > So is Andrew's machine an ingest machine?
> >
> > but instead I guess I'll just request the data
> > > like shemp is now. But, this confuses me further. Is it that our
> > > wsihcsn is not the way it came from WSI?
> > >
> > It's the same type of configuration as the NOAAport ingest machine desi,
> > huh. Think ingest(source) verses downstream.
> >
> > Robb...
> >
> > > Thanks for your help!
> > >
> > > Anne
> > >
> >
> > ===============================================================================
> > Robb Kambic Unidata Program Center
> > Software Engineer III Univ. Corp for Atmospheric
> > Research
> > address@hidden WWW: http://www.unidata.ucar.edu/
> > ===============================================================================
>
> Hi Robb,
>
> From our discussion yesterday, I thought that our wsihcsn was an ingest
> machine that came from WSI that ran their software, thus requiring
> 'accept' and 'allow' on wsihcsn.
Anne,
That is true because they are running an older version of the LDM
requiring an accept. But once the WSI products are entered into the queue
it's standard LDM configurations are only needed. Your confusion about
the accept is that sites getting products from WSI need to have an accept
because WSI wrote their own s/w to distribute the products, it's not an
"LDM". This configuration requires an accept.
There are two different sources for the WSI data, an ingest machine like
our wsihcsn and WSI sending products over the internet to remote sites.
But, once the data is in the queue only the standard request line is
needed for the downstream nodes.
that wanted to feed from it, such as
> shemp.
So, I thought it was different than requesting from desi. This
> is what I assumed was the case for Andrew as well. Indeed, in Andrew's
> ldmd.conf he has the 'accept' and 'allow' lines:
>
> allow WSI wsihcsn
> accept WSI .* ^wsihcsn
>
> So, this particular host of Andrew's is not an ingest machine. I'm
> assuming his wsihcsn *is* an ingest machine.
>
That's what I thought since the short hostname.
> But, our wsihcsn is running some version called ldm-5. It's not like
> our usual distribution. The src dir points to /upc/share/ldm/src, which
> doesn't exist. It also has a subdir called 'build'. In 'build' there
> are files
Don't worry about that, it was only done to create LDM builds on Solaris
5.6. wsihcsn is the only 5.6 machine at Unidata.
RObb...
>
> lrwxrwxrwx 1 ldm ldm 16 Oct 27 1998 COPYRIGHT ->
> ../src/COPYRIGHT
> lrwxrwxrwx 1 ldm ldm 14 Oct 27 1998 INSTALL ->
> ../src/INSTALL
> lrwxrwxrwx 1 ldm ldm 15 Oct 27 1998 MANIFEST ->
> ../src/MANIFEST
> lrwxrwxrwx 1 ldm ldm 18 Oct 27 1998 Makefile.in ->
> ../src/Makefile.in
> lrwxrwxrwx 1 ldm ldm 13 Oct 27 1998 README ->
> ../src/README
> lrwxrwxrwx 1 ldm ldm 20 Oct 27 1998 RELEASE_NOTES ->
> ../src/RELEASE_NOTES
> lrwxrwxrwx 1 ldm ldm 14 Oct 27 1998 VERSION ->
> ../src/VERSION
> lrwxrwxrwx 1 ldm ldm 17 Oct 27 1998 aclocal.m4 ->
> ../src/aclocal.m4
> lrwxrwxrwx 1 ldm ldm 16 Oct 27 1998 configure ->
> ../src/configure
> lrwxrwxrwx 1 ldm ldm 19 Oct 27 1998 configure.in ->
> ../src/configure.in
> lrwxrwxrwx 1 ldm ldm 21 Oct 27 1998 ldmconfig.h.in ->
> ../src/ldmconfig.h.in
>
> but none of those ../src/* files exist.
>
> So, what software is our wsihcsn running? Ours or theirs? If it's
> theirs, then I'm surprised to see the 'request' line on shemp.
>
> Anne
>
===============================================================================
Robb Kambic Unidata Program Center
Software Engineer III Univ. Corp for Atmospheric Research
address@hidden WWW: http://www.unidata.ucar.edu/
===============================================================================