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

[ldmMcidas #FNK-457904]: Persistent SATANNOT/SATBAND error message.

Hi David,

Sorry for the tardy reply...

re:
> Thanks for the welcome!

It is always nice to welcome a new member of the community (you, not Creighton)!

re:
> As a note...we are not running MCIDAS on this
> system (from what I've been told), although it is something that has been
> done in the past...in order to produce wxp output files.

Yes, Creighton used to run McIDAS, but not to output WXP files.  Presumably
WXP was run to do that.  That is not important for this discussion, however.

re:
> After doing a bit
> more mucking around in the code today, I am on the same page...the
> EAST-CONUS files are not part of the UNIWISC feed (they are NIMAGE), so
> that's another story entirely!

Yup.

re:
> As an aside, I have isolated the SATANNOT
> and SATBAND errors to the following 5 sets of feeds...
> 
> GOES11 - 4km/IR
> GOES11 - 8km/WV
> GOES13 - 4km/VIS
> GOES13 - 4km/WV
> GOES13 - 4km/IR

Hmm...  I could understand there being a warning issued when the platform
is GOES-13 _if_ the SATANNOT and SATBAND files you have are very old.  I
guess if they are old enough, then GOES-11 wouldn't be recognized either.

The solution to this problems should be you loading in a more current
version of the ldm-mcidas decoders, or, at least, the SATANNOT and
SATBAND files from a more current version of the decoders (this may
be tricky, however, since the format of the SATBAND file changed some
time ago, and the code in pnga2area had to change to accommodate the
change).  You can download a binary distribution of ldm-mcidas v2008
using anonymous FTP as follows:

machine:   ftp.unidata.ucar.edu
user:      anonymous
pass:      address@hidden
directory: pub/binary/linux_2.6-i686
           OR
           pub/binary/linux_2.6-x86_64
file:      ldm-mcidas-2008.tar.gz

I would download this package to a directory under ~ldm:

<as 'ldm'>
cd ~ldm
mkdir ldm-mcidas
cd ldm-mcidas
ftp ftp.unidta.ucar.edu
  user: anonymous
  pass: address@hidden
  cd pub/binary/linux_2.6-i686
  binary
  get ldm-mcidas-2008.tar.gz
  quit
tar xvzf ldm-mcidas-2008.tar.gz
rm -f runtime
ln -s ldm-mcidas-2008 runtime
ln -s runtime/* .

At this point, you could try replacing the copies of SATANNOT and SATBAND
that you have in ~ldm/etc with the ones in the v2008 distribution:

cp etc/SAT* ~/etc

NB: if your ldm-mcidas decoders are _really_ old, your copy of pnga2area
may not recognize the different format in SATBAND.  In this case, I would
verify that the v2008 binaries will run on your machine, and if they
do copy them over to the directory where your current ldm-mcidas decoders
live:

ldd bin/pnga2area
ldd bin/nldn2md
ldd bin/lgt2md
etc.

If the listing from 'ldd' shows that all needed shared libraries are
found, then copy the decoder(s) you are using.  For instance,

cp bin/pnga2area ~/decoders      <- assumes that your existing copy of
                                    pnga2area is in ~ldm/decoders

However, please read on... you may not need to do this at all.

re:
> Here are the pqact.conf and ldmd.conf files that you were asking for.  I
> have also attached the pqact.gempak_images as well...this contains the other
> set of pnga2area calls (without the -a and -b flags).

OK, thanks for sending these along.  Here are two quick observations:

- you are sending all of the products for all of the datastreams you are
  ingesting to be processed by the actions in ~ldm/etc/pqact.conf

  It is highly likely that you really don't want to be doing this.  There
  are two reasons I say this:

  - there are only actions for UNIWISC (aka MCIDAS) and NIMAGE products
    in ~ldm/etc/pqact.conf

  - you are processing the products from the UNIWISC (aka MCIDAS) and
    NIMAGE feeds with actions in the ~ldm/etc/pqact.gempak_images file

  Since you note that you are not running McIDAS, I would think that
  you do not need to do any of the processing embodied in the
  ~ldm/etc/pqact.conf file.  If this is true, all you need to do is
  comment out the line in ~ldm/etc/ldmd.conf that runs those actions:

  change:

  exec    "pqact"

  to:

  #exec    "pqact"

  NB: any time a change is made in ~ldm/ldmd.conf, you have to stop/start
  the LDM for it/them to take effect:

  ldmadmin restart

- the MCIDAS and NIMAGE actions embodied in ~ldm/etc/pqact.gempak_images
  are FILE actions only; they do not run the ldm-mcidas 'pnga2area' decoder

  This means three things:

  - these actions can not cause the SATANNOT/SATBAND warnings you are seeing

  - if GEMPAK processing is all you need/want to do, then commenting out the
    exec    "pqact" line from ~ldm/etc/ldmd.conf and restarting the LDM
    should stop all of the SATANNOT/SATBAND warning messages you are seeing

  - there should be no reason for your not being able to view EAST-CONUS images
    in GEMPAK applications

    Question:

    - did you verify that you have images in the 
~ldm/data/gempak/images/sat/EAST-CONUS/4km/*
      directories?

      If yes, the inability to view the images in GEMPAK is a mystery to me.

re:
> The paths are all set
> properly and the files are in the ~/etc/ directory.  So that's that!

Yup.

re:
> One
> more (hopefully) simple question.  I can't find anything that describes the
> difference between UNIWISC and MCIDAS references in the *.conf files...do
> they call/mean the same thing?

Yes, they are the same thing.  For years the datafeed was known as MCIDAS
since it was the primary mechanism for delivering satellite imagery to
the McIDAS application.  However, since GEMPAK, McIDAS and the IDV can
all use the imagery in the datastream, it was decided to rename it (actually
to make its primary name) to UNIWISC since this more closely reflects
the actual name of the datastream, the Unidata-Wisconsin datastream.  You
will note that when I reference UNIWISC, I always add '(aka MCIDAS)'.  This
is to alert the user that UNIWISC and MCIDAS are synonymous.

re:
> Regardless, thanks for the help!

No worries.  We are always happy to help when we can.

re:
> I just
> moved here to Creighton and Omaha after living my whole life in Denver, so
> there are a lot of changes here!

Now that is an interesting move...  How do you like it?

re:
> Hope all is well back home...all the best!

Things are very nice in Colorado of late.  Lots of nice fluffy snow (and cold)
followed by warm (but windy) weather.  It was so warm this past weekend that
we had our doors and windows open in my house which is located at 8150'
on Sugarloaf!

Cheers,

Tom
--
****************************************************************************
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: FNK-457904
Department: Support ldm-mcidas
Priority: Normal
Status: Closed