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

[LDM #HMA-788722]: LDM problem

John,

> Well, NEXRAD2 is coming in nicely now.  However, I see CONDUIT
> latencies are all over the map:
> 
> http://www.unidata.ucar.edu/cgi-bin/rtstats/iddstats_nc?CONDUIT+adiabat.rwic.und.edu
> 
> I know for sure we're not getting all the RUC products.  Sigh.
> 
> I also spoke with some folk in the network spaces.  They swear all up
> the chain of command that there is no traffic shaping going on on our
> I2 links.  We do share our I2 pipe with some other universities in the
> state, so I'm trying to see if there's anything going on on the state
> level.  What fun.

I might have misled you :-(  Colleague Tom Yoksas (who's been monitoring this
exchange) pointed-out that most of your smaller-volume feeds are coming from 
marconi.rwic.und.edu and not from us.  The exception is the GPS feed, whose
interpretation is problematical.  As a consequence, we can't say for certain
that the flow from us to you is being shaped.  :-(

Because the CONDUIT feed has greater volume than the NEXRAD2 feed, if
packet shaping were occurring, then that feed would also be affected.

I suggest trying an experiment.  Replace the REQUEST for IDS|DDPLUS data
from Marconi with one to idd.unidata.ucar.edu.  That way you'll get
both high-volume and low-volume feeds from us and we'll be in a much
better position to interpret the results.
Regards,
Steve Emmerson

Ticket Details
===================
Ticket ID: HMA-788722
Department: Support LDM
Priority: Normal
Status: On Hold