[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: 20000508: no ldmstats for NSBF on FOS page

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: Re: 20000508: no ldmstats for NSBF on FOS page
  • Date: Mon, 8 May 2000 12:47:34 -0600 (MDT)
On Mon, 8 May 2000, Unidata Support wrote:

> >To: address@hidden
> >From: weather <address@hidden>
> >Subject: no ldmstats for NSBF on FOS page
> >Organization: UCAR/Unidata
> >Keywords: 200005081626.e48GQGG06787
> 
> I was looking at the FOS Topology page
> 
> http://www.unidata.ucar.edu/projects/idd/status/idd/fosTopo.html
> 
> and I used to see our statistics on there for wxmcidas and psnldm.
> 
> There names are fully qualified and they used to appear there,
> but now are gone.  The stats are beinf produced.  Are you not
> receiving them?

Robert,

Yes UPC is receiving your stats from both psnldm/wxmcidas. The entries
from the topology pages were missing because I flushed out the topo pages
on Fri to get rid of old wrong entries and the new links where not
generated immediately.  I entered the link from thelma to cirrus so your
entries should now appear.


I'm concerned on your configuration for psnldm/wxmcidas.  It is UPC policy
to only feed one machine on a site from an upsteam site because the
limited number of feeder sites and possibility of network congestion
caused by receiving duplicate products.  The duplicate product detection
is done on the downstream site, so the duplicate product is actually sent
to the downstream node most of the time before it's discarded. This
configuration might result in worst performance than better. I would like
you to only use one machine.  There are other options to get higher
data reception.  Split the feed types up, using the IP number for some of
the feed types and the fully qualified hostname for the other.  This
generates a receiver for the IP and the hostname requests. Also, you
can install ldmfail script to check the status of the upstream nodes.
There is a man page on how to configure it. These options are more
desirable IDD  type of configurations. 

Robb... 





> 
> Thanks,
> Robert Mullenax
> 
> cirrus.al.noaa.gov is my feed and psnldm/wxmcidas feed each other
> 

===============================================================================
Robb Kambic                                Unidata Program Center
Software Engineer III                      Univ. Corp for Atmospheric Research
address@hidden             WWW: http://www.unidata.ucar.edu/
===============================================================================