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

20000915: McIDAS-XCD decoding and McIDAS updates at UNL



>From: Clint Rowe <address@hidden>
>Organization: University of Nebraska-Lincoln
>Keywords: 200009151802.e8FI2Ob09947 McIDAS-XCD decode LDM pqact xcd_run

Clint,

>Mark came to me this morning and told me that we haven't been getting
>any data into McIDAS all semester (we're in the third week now, so
>apparently it wasn't too important).

I hate to hear comments like this :-(

>We are receiving data and it's
>getting into gempak files without any noticable problems, so the problem is 
>somewhere in McIDAS.  AREA files seem to be okay, as do the profiler data.

With the logins you provided, I found that someone had deleted the XCD
actions from ~ldm/etc/pqact.conf:

#########################################################################
#
# McIDAS-XCD section
#
DDPLUS|IDS      ^.*     PIPE
        xcd_run DDS
HRS     ^.*     PIPE
        xcd_run HRS

The data was, therefore, not going to McIDAS-XCD spool files and so
was not being decoded.  Now, who could have deleted these entries ;-)?

>When I went to start McIDAS as user mcidas I got the following:
>
>Error in startup script: can't read "str(?GFILE)": no such variable
>    while executing
>"upcRunCommand "IGU DEL $str(?GFILE);IGU MAKE $str(?GFILE) 2000" """
>    (file "/home/mcidas/bin/mcgui.k" line 1337)

This is a separate problem that has nothing to do with McIDAS-XCD
decoding.

>Anyway, something more than just data problems seems to be happening
>and I don't know how long it's been going on -- I haven't really done
>anything since I installed the png decoder mid-summer.  I'll be glad
>to give you the login(s) if you want to poke around.  In fact, to save 
>time, here they are -- have at it!

OK, after a few minutes of waiting after I put the xcd_run lines back in
pqact, data files are once again being decoded.

Interestingly, a long time-based listing in /data/mcidasd is showing
that there is probably something wrong with your system clock:

ls -alt | more
total 633738
-rw-rw-r--   1 ldm      apps     1016480 Sep 15  2000 DD002590.XCD
-rw-rw-r--   1 ldm      apps     16781313 Sep 15  2000 HRS.SPL
-rw-rw-r--   1 ldm      apps     4924864 Sep 15  2000 MDXX0069
-rw-rw-r--   1 ldm      apps     26481036 Sep 15  2000 MDXX0009
-rw-rw-r--   1 ldm      apps        8192 Sep 15  2000 SM00259.IDX
-rw-rw-r--   1 ldm      apps       12288 Sep 15  2000 SA00259.IDX
-rw-rw-r--   1 ldm      apps        3140 Sep 15  2000 ZZ00259.IDX
-rw-rw-r--   1 ldm      apps        5120 Sep 15  2000 MET00259.IDX
-rw-rw-r--   1 ldm      apps        1600 Sep 15  2000 SPE00259.IDX
-rw-rw-r--   1 ldm      apps        1220 Sep 15  2000 UA00259.IDX

These files are newly updated.  Notice how there is no time associated with
them.

>Let me know what you find.

On the McIDAS side, I see that you are running the original 7.600 release
with no bugfixes installed.  I am taking the liberty of updating the
McIDAS installation to current bugfix levels (7.613) (this appears
like it is going to take a long time, so I will send this email before
the job is done).

<as 'ldm'>
cd decoders/bin
<edit mcscour.sh and change the range of GRID scouring to 5001 - 6000>

<as 'mcidas'>
cd mcidas7.6/update
ftp ftp.unidata.ucar.edu
  <user> umcidas
  <pass> XXXXXX
  binary
  cd unix/760/bugfix
  get mcupdate.tar.Z
  quit
cd ../src
touch Mcmkmcgrid.c
make all
make install.all

Other changes I will make to the 'mcidas' account are:

o making sure that your ADDE setup is correct (as far as datasets goes)
o making sure that file REDIRECTions are correct/usable (I have already
  made a couple of these)
o finding and fixing the problem you sent above (if the updates don't
  already fix it)
o anything else (within reason :-) you ask for

I noticed that you did not have an invocation of mcscour.sh in either
'ldm's or 'mcidas' crontab files.  I added one to 'ldm's:
#
# Scour McIDAS XCD data files
#
00 21 * * * decoders/bin/mcscour.sh

>Thanks in advance,

So, guess who the first person to note that the delta-encoded images
were missing from the Unidata-Wisconsin datastream (they were removed
as of 18Z)...  Harry, of all people! :-)

Tom