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

20051004: IDS DDS problem?



>From:  =?ISO-8859-1?Q?Christian_Pag=E9?= <address@hidden>
>Organization:  UQAM
>Keywords:  200510041834.j94IYLG7018300 IDD 

Hi Christian,

I know that Chiz has already replied to you on your IDS|DDPLUS latency
problem, but there is more to the story.  Please pay particular
attention to the last two paragraphs of this note.  Thanks.

re:
>I dont't receive anymore IDS/DDS on my LDM. Is there a problem?

There is no problem that we can see with the receipt of IDS|DDPLUS data
through NOAAPORT.  I do notice, however, that the latency being seen on
your IDD node, io.sca.uqam.ca, spiked at about 14-15 UTC:

http://www.unidata.ucar.edu/cgi-bin/rtstats/iddstats_nc?IDS|DDPLUS+io.sca.uqam.ca

This plot shows very high latencies to both of your upstream feed
hosts:

flood.atmos.uiuc.edu
omega.lsc.vsc.edu

The dramatic increase in latency from the UIUC relay is apparently a
result of its receipt of the IDS|DDPLUS data:

http://www.unidata.ucar.edu/cgi-bin/rtstats/iddstats_nc?IDS|DDPLUS+flood.atmos.uiuc.edu

The latest rtstats data coming from UIUC, however, shows that the
latencies are now dropping rapidly.  Hopefully, your latencies will
follow the UIUC's trend.

If you find that you continue to have problems, you can switch your
IDS|DDPLUS feed request to atm.geo.nsf.gov.  Your IDS|DDPLUS reception
from atm should be good since the latencies you are seeing in the much
higher volume NIMAGE data from atm look acceptable:

http://www.unidata.ucar.edu/cgi-bin/rtstats/iddstats_nc?NIMAGE+io.sca.uqam.ca

One last comment:  the volume of IDS|DDPLUS data that you have been
ingesting from the Lyndon State relay node, omega.lsc.vsc.edu, is
ranging up to ten times the volume that it should be:

http://www.unidata.ucar.edu/cgi-bin/rtstats/iddstats_vol_nc?IDS|DDPLUS+io.sca.uqam.ca
http://www.unidata.ucar.edu/cgi-bin/rtstats/iddstats_vol_nc?IDS|DDPLUS+omega.lsc.vsc.edu

Since the Lyndon State relay node is getting IDS|DDPLUS data from another
Lyndon machine, we believe that they are running a NOAAPORT ingest system
that is putting data products into the IDS|DDPLUS stream in a manner
that is inconsistent with the operation of the IDD.  For this reason, we
suggest you discontinue your IDS|DDPLUS feed from Lyndon at least until
we can sort out what is happening.

Cheers,

Tom
--
+-----------------------------------------------------------------------------+
* Tom Yoksas                                             UCAR Unidata Program *
* (303) 497-8642 (last resort)                                  P.O. Box 3000 *
* address@hidden                                   Boulder, CO 80307 *
* Unidata WWW Service                             http://www.unidata.ucar.edu/*
+-----------------------------------------------------------------------------+