20020410: McIDAS decoding questions

>From: William C Klein <address@hidden>
>Organization: Valparaiso
>Keywords: 200204101351.g3ADpma28875 McIDAS-XCD ldm-mcidas

Bill,

>I have a few questions.  We seemed to have gotten most of our issues
>straightened out, but have just a few more.
>
>1.  Certain datasets are not being picked up:
>       -NLDN (we have been granted access to this, but not getting)

There are several steps in the process of making NLDN lightning data
available in McIDAS:

o get permission for a point-to-point LDM feed of the data from SUNY
  Albany

o setup your ~ldm/etc/ldmd.conf to actually request the data from SUNY Albany

o setup your ~ldm/etc/pqact.conf file to run the ldm-mcidas decoder
  to decode the NLDN data received from SUNY Albany.  In order for this
  step to work, you will have and to copy the SCHEMA, ROUTE.SYS,
  and SYSKEY.TAB files from the McIDAS-X distribution to the directory
  in which the NLDN MD files are to be written/read.  Also these
  files must have the correct read/write permission.  Here is a quick
  howto:

  <login as 'mcidas'>
  cd workdata
  cp ROUTE.SYS /data/mcidas            <- assumes you are decoding data into
                                          the /data/mcidas directory; adjust
                                          as necessary
  cd ~/data
  cp SCHEMA /data/mcidas
  cp SYSKEY.TAB /data/mcidas
  cd /data/mcidas
  chmod 664 ROUTE.SYS SYSKEY.TAB       <- assumes 'ldm' and 'mcidas' are in
                                          the same group; if not use 666
  chmod 666 SCHEMA
  
o you setup appropriate file REDIRECTions so that your 'mcidas' account
  can find the files into which the NLDN lightning data is to be decoded

o you setup the RTPTSRC dataset in the 'mcidas' account to access the
  decoded NLDN data

All of this will seem pretty complicated until you have done it a couple
of times, and then it becomes easy.  Please let me know if one or more
of the steps above was not done or if they were done and things stil
are not working.
  
>       -NOWRAD

The NOWrad (tm WSI Corp) data has to be purchased from WSI.  It is likely
that you did not do this, so you will not be able to look at those
composite NEXRAD products.  The as yet unannonced news on this front,
however, is that we are creating several composite radar reflectivity
products that are now available for free.  To get started on looking
at these products in McIDAS, you will need to download and install the
newest McIDAS addendum (bugfix).  After doing this, you will no longer
have the RTNOWRAD dataset as a default option in your MCGUI Image display
list.  Instead, you will have a new dataset called NEXRCOMP.  NEXRCOMP
has 4 different products:

10 km National Radar Coded Message (RCM) composite reflectifity
6 km National NEXRAD Level III base reflectivity (N0R) composite
1 km Regional (floater) NEXRAD Level III composite
1 km National NEXRAD Level III base reflectivity composite

All of these products are currently available from a set of cooperating
Unidata community ADDE servers:

pscwx.plymouth.edu
atm.geo.nsf.gov
cacimbo.ggy.uga.edu
papagayo.unl.edu
stratus.al.noaa.gov  (may change in the coming days)
adde.ucar.edu

The steps for upgrading your McIDAS distribution are:

1) login as 'mcidas'

2) cd mcidas7.8/update

3) FTP the addendum to your machine and unpack it:
   ftp ftp.unidata.ucar.edu
     <user> umcidas
     <pass> XXXXXX
     cd unix/780/bugfix
     binary
     get mcupdate.tar.Z
     quit
   ./mcunpack

4) rebuild your distribution using the new code:

   cd ~mcidas/mcidas7.8/src
   make all

5) shutdown your LDM for a few minutes while you install the new
   McIDAS code:

   <login as 'ldm'
   ldmadmin stop

6) install the new McIDAS code

   <back as the user 'mcidas'>
   cd ~mcidas/mcidas7.8/src
   make install.all

7) after the McIDAS installation finishes, restart your LDM:

   <as 'ldm' again>
   ldmadmin start

After doing this, exit any/all McIDAS sessions.  Then, start a new
McIDAS session that uses the MCGUI and click on the ADDE Client Routing
Configuration button on the MCGUI (the button just to the right of the
button with the Z on it).  Select the Images tab and click on the down
arrow to the right of the NEXRCOMP line and select a server from the
above list to point to for NEXRCOMP data.

>       -NCEP forcast grids

This is most likely not working because the McIDAS-XCD GRID decoding 
has never been turned on, or there is no appropriate action in
the LDM pqact.conf file that would send the model data to the
spool that the McIDAS-XCD GRID decoder uses.  Do the following
as the user 'mcidas':

<login as 'mcidas'>
cd workdata
decinfo.k LIST

See if the listing from decinfo.k shows that DMGRID is 'active' or
'inactive'.  If it is inactive, run the following:

decinfo.k SET DMGRID ACTIVE

Now, the GRID files produced by the McIDAS-XCD grid decoder are
HUGE.  With the latest code modifications in the addendum that I
have you install above, you will need several GB of disk just for
GRID decoding.  Exactly how much disk you will need will depend on
which grids you decided to decode (this is configurable) and how
many days you want to keep online at the same time.

>       -FOUS14 forcasts
>    We can see that we get the sets under the Data Availability.

The FOUS14 decoding should be working as it is turned on by default
in the XCD setup I send out, and the fact that you can see sets
under Data Availability means that you should be able to view the
data.  What are the messages that are being issued when you try to
plot/contour the data?  To see this, open up the MCGUI Command
window (click the left button on the keyboard icon (to the left of
the button with the Z) and look at the command that was run
from the MCGUI when you tried to display FOUS14 data.

>2.  How can we setup a data-fail-over to a different site if our provider
>goes down?

If you are referring to LDM-delivered data, then you need to have
a secondary site from which you can feed data (you should have already
been assigned such a site when you first got an LDM feed site).

Our records indicate that you should be able to feed from either
weather.cod.edu or ldm.meteor.psu.edu.

After you have a primary and failover feed assigned to you, and after
making sure that both sites will feed you, you can setup a cron action
in your 'ldm' account that will periodically kick off and see if the
LDM you should be feeding from is operational.  If it is not, the
LDM is shutdown and restarted for the failover site.  The script
that will be run out of cron is 'ldmfail', and information on how
to set this up is continaed in the online LDM pages on our site.

Please let me know if I can help get you going on any of the above items.

Tom Yoksas