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

20050124: Unidata McIDAS-XCD Trouble (cont.)



>From: "James T. Brown" <address@hidden>
>Organization: MSU
>Keywords: 200501201933.j0KJXOv2020815 McIDAS-XCD ADDE

Hi Jim,

>Some good news and some bad news.

Hmm...

>1) The good news - I built a new install of McIDAS and the XCD decoders
>on the LDM machine (accas).

I did this as well on Sunday night.

>Before activiting, I commented out the
>NOAAPort feed from the "ldmd.conf" file just to see if it would make a
>difference.   The MD files seem to be processing normally now.

Super!

>The
>next step would be to turn the NOAAPort feed back on using the regular
>expression you list below.

Yup.  I am interested in the results too.

>2) The bad news - I now have two installs.  One on PILEUS which seems
>to work and a local install on ACCAS, but I may have messed something
>up or there is a problem with the firewall.  I can't access any of the
>LOCAL-DATA or the new MDxx files that are coming in using the ACCAS
>built:
>
>   ptlist.k RTPTSRC/SFHOURLY
>
>produces output when run on the PILEUS build, but not the ACCAS build.
>ACCAS returns the following error:
>
>   accas% ptlist.k RTPTSRC/SFHOURLY
>   ptlist.k: Point data server unable to resolve this dataset: 
>RTPTSRC/SFHOURLY

You must have changed something because I just tried this as 'mcidas'
on accas and it worked as it should.

>I have all my notes from the two builds and tried to do everything the
>same.  Even when I disable the firewall, still no good on ACCAS.

If they were separate builds, the datasets would need to be setup
in the same way as on pileus:

- make a copy of DSSERVE.BAT as LSSERVE.BAT
- edit LSSERVE.BAT to set values specific to your install
- define the McIDAS string XCDDATA to be /data/mcidas
- run 'batch.k LSSERVE.BAT'

At this point the datasets should be defined.

If you want to use the remote ADDE interface on accas, you also need
to do the following:

<as 'root'>
-- create the user 'mcadde' to have the same HOME directory as 'mcidas'
   but not have a login shell

cd /home/mcidas
sh ./mcinet2004.sh install mcadde

<as 'mcidas'>
create /home/mcidas/.mcenv     just as you did on pileus

At this point the remote ADDE access should work on accas.

Now, since I "pointed" at both pileus and accas from my McIDAS session
here at the UPC and I was able to access data from the RTPTSRC dataset,
I will assume that you have configured everything to work.

>Eventually, I am going to try to run the ADDE services on the LDM
>machine - eventually!

It seems to work now:

DATALOC ADD RTPTSRC ACCAS.GEO.MSU.EDU

Group Name                    Server IP Address
--------------------         ----------------------------------------
RTPTSRC                      ACCAS.GEO.MSU.EDU

<LOCAL-DATA> indicates that data will be accessed from the local data directory.
DATALOC -- done
 
PTLIST RTPTSRC/PTSRCS FORM=FILE ALL
Pos      Description                        Schema  NRows NCols Proj# Created 
DataDate
------   --------------------------------   ------  ----- ----- ----- ------- 
--------
     4   SAO/METAR data for   24 JAN 2005   ISFC       72  7000     0 2005025 
2005024
     5   SAO/METAR data for   25 JAN 2005   ISFC       72  7000     0 2005025 
2005025
    14   Mand. Level RAOB for 24 JAN 2005   IRAB        8  1500     0 2005025 
2005024
    15   Mand. Level RAOB for 25 JAN 2005   IRAB        8  1500     0 2005025 
2005025
    16   Mand. Level RAOB for 26 JAN 2005   IRAB        8  1500     0 2005025 
2005026
    24   Sig.  Level RAOB for 24 JAN 2005   IRSG       16  6000     0 2005025 
2005024
    25   Sig.  Level RAOB for 25 JAN 2005   IRSG       16  6000     0 2005025 
2005025
    26   Sig.  Level RAOB for 26 JAN 2005   IRSG       16  6000     0 2005025 
2005026
    34   SHIP/BUOY data for   24 JAN 2005   ISHP       24  2000     0 2005025 
2005024
    35   SHIP/BUOY data for   25 JAN 2005   ISHP       24  2000     0 2005025 
2005025
    45   NGM MOS for day      25 JAN 2005   FO14       38   600     0 2005025 
2005025
    53   SYNOPTIC data for    23 JAN 2005   SYN         8  6500     0 2005025 
2005023
    54   SYNOPTIC data for    24 JAN 2005   SYN         8  6500     0 2005025 
2005024
    55   SYNOPTIC data for    25 JAN 2005   SYN         8  6500     0 2005025 
2005025
    56   SYNOPTIC data for    26 JAN 2005   SYN         8  6500     0 2005025 
2005026
    64   PIREP/AIREP data for 24 JAN 2005   PIRP       24  1500     0 2005025 
2005024
    65   PIREP/AIREP data for 25 JAN 2005   PIRP       24  1500     0 2005025 
2005025
   105   ETA MOS for day      25 JAN 2005   ETA        42  2000     0 2005025 
2005025
   115   GFS MOS for day      25 JAN 2005   GFS        42  2000     0 2005025 
2005025
PTLIST: Done

>This is a new problem and I don't intend to have
>to bother you with it, especially if I the original install on PILEUS
>will work - at least that ADDE server seems to be processing data
>properly so I guess some progress has been made.

Looks OK to me.  I note that not all datasets are setup on accas, however.
By the way, the settings made when you run 'batch.k LSSERVE.BAT' get
saved in the file ~mcidas/workdata/RESOLV.SRV.  Knowledgable users can
edit this file directly to shortcut the ADDE dataset creation step.
I have to warn you, however, that it is very easy to make a typo and
have ADDE access stop working.  I know, 'cause I have done this with
something as stupid as deleting a trailing ',' (comma) on a single
RESOLV.SRV line.  When this happens, dataset definitions after this
point will not work.

>You asked what type of NOAAPort system we have.  It is from UNISYS and
>we are not receiving all of the channels.  We only subscribe to the
>NWSTG channel which I believe is just text - we are not receiving any
>satellite or gridded data via NOAAPort.

The NWSTG channel contains textual products (the contents of the
IDS|DDPLUS datastream), model output (the contents of the HDS datastream)
and NEXRAD Level III products (the contents of the NNEXRAD datastream.

>Feel free to log back in to take a look if you wish, especially the
>build on ACCAS (however, do not feel obligate to be working on this
>until 2:30am!

I logged back on a few minutes ago and see that everything is working
nicely now.  Like I said above, however, not all datasets have been
defined on accas.  The only ones that have been defined are TOPO
and RTPTSRC.  Also, no worries about the late night session on Sunday.
I got online from my house and was taking care of other pending email
when I decided to jump onto pileus to take a closer look.

>It is 8:45pm out here and that is about late enough for
>me - I need to get out of the office and clear my head for a bit and
>will tackle this some more in the morning).

I'm with you there :-)

>Thanks for all your help!

No worries.  I'm glad that decoding is working now.  I will be
very interested in seeing if the regular expression I sent
earlier will be useful in decoding data from your NOAAPORT system.

Speaking of your NOAAPORT system... are you planning for what to
do when the broadcast switches to DVB-S (Digital Video Broadcast -
Satellite)?  This is a hot topic at the moment...

Cheers,

Tom
--
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.