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

[McIDAS #EHH-480032]: LDM filtering

Hi Paul,

> Actually I'm using the 2 year old noaaport software (noaaport-1.0.8) and
> (ldm 6.2.1) on my dvb receiver box and on my down stream ldm with xcd I
> have (ldm 6.3.0) This is my non operations Linux test platform so that's
> why I installed a newer version of LDM.

I strongly suggest that you upgrade your NOAAPort/LDM and downstream LDM:

- the latest NOAAPort ingest package is available by anonymous FTP from
  the Unidata FTP site, ftp.unidata.ucar.edu:

  <login as 'ldm'>
  cd ~ldm
  ftp ftp.unidata.ucar.edu
    <user> anonymous
    <pass> address@hidden
    cd pub/noaaport
    get noaaport-1.3.0.tar.gz
    cd ../ldm
    get ldm-6.6.5.tar.gz


  1) The newer versions of the NOAAPORT code _require_ newer (6.4+)
     versions of the LDM.  They will not link against older versions
     (pre 6.4)!!!!!

  2) The new LDM must be completely installed before building the
     new NOAAPORT code.  This includes changing the runtime link in
     the ~ldm directory to point at the new LDM installation.  The
     reason for this is that there were changes in the LDM that required
     changes in the NOAAPORT code.

- you can upgrade the LDM on your downstream machine whenever you like,
  but you must do the upgrade if you want the ability to control the
  data sent to a downstream from the upstream LDM.  The point is that
  your original request for how to craft a regular expression that would
  request just the data you want can be turned around to ask how the
  upstream can limit what the downstream can request.  This upstream
  filtering ability was introduced in LDM-6.4+.

> On the dvb receiver, in ldmd.conf I uncommented the "dvbs_multicast" and
> "readnoaaport" entries with IP's, .3 and .4 (six lines total).
> Then restarted ldm. Each IP corresponds to an old NOAAPort channel 1-4
> and dvb pid 101-104. I believe pid 103 is the NWSTG2.

Very good.  I still recommend that you upgrade the NOAAPort software on
the DVBS ingest machine since there have been numerous modifications to
the code that enable you to receive more data more reliably and better
classified into the various datastreams.

> On the down stream ldm I added lines to:
> ldmd.conf:
> request NGRID ".*" <HOST> PRIMARY
> pqact.conf:
> NGRID ^.* PIPE /home/oper/mcidas/bin/ingebin.k NBIN
> Then restarted ldm.

OK.  The regular expression you need to get just the data you want
from the NGRID datafeed is what you are looking for.  The problem with
me trying to craft one for you to use is that I will be looking at
the NGRID feed as it is created by the most recent NOAAPort ingest
code, and you will not UNLESS you upgrade.

> Another bug not related to NOGRIB is a scaling issue with MSL pressure.
> The grib2 file has units of PA with a scale of 0. When I tried to plot
> the grid over a SE US sector I had to use a CINT=60 for pressure to get
> it to plot and the contour labels showed a six digit integer. The
> workaround for this is to use POWER=-2 when plotting MSL pressures. I
> don't know if this is for all MSL pressure grib2 files or just the
> GFS-USLC2 which is what I was working with.

I had not noticed this problem.  I will take a look later today.

One problem I saw right off was the improper unpacking of GRIB2 messages
for situations when the scanning mode in the GRIB message is back and
forth across columns and then rows.  The GRIB2 messages that have this
scanning result in a plot where every other line is plotted in reverse

> Another issue I had was plotting a parameter and letting it default to
> the domain of the grid. When I tried this it didn't plot anything. I
> still need to check this in combination with the POWER keyword to see if
> I can get it to plot anything.

If/when you run into these fields, please let me know so I can take a look.

> For now at least yes, the Alaska projection products are what I want to
> eliminate. There will be more once I figure out how to do it. If I
> create a regular expression for these grib2 files won't that be all I
> get? Is there a way to define a "NOT" regular expression. Of course I'll
> be the first to admit I'm an R.E. neophyte so it doesn't mean you can't
> do it, I just haven't seen it done anywhere.

OK.  With LDM 6.4+ you can do some/all of the work on the upstream side and
some/all of the work on the downstream side.  Which approach is best is
unclear to me at this moment.  I will consult with Steve Chiswell, our
GEMPAK and model expert, to see what guidance he can provide.


Unidata User Support                                    UCAR Unidata Program
(303) 497-8642                                                 P.O. Box 3000
address@hidden                                   Boulder, CO 80307
Unidata HomePage                       http://www.unidata.ucar.edu

Ticket Details
Ticket ID: EHH-480032
Department: Support McIDAS
Priority: Normal
Status: Open

NOTE: All email exchanges with Unidata User Support are recorded in the Unidata inquiry tracking system and then made publicly available through the web. If you do not want to have your interactions made available in this way, you must let us know in each email you send to us.