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

20011108: IMGDISP problems - workaround - (cont.)



>From: Mike Voss <address@hidden>
>Organization: SJSU
>Keywords: 200110181917.f9IJHA119842 McIDAS IMGDISP ADDE MAP comment card

Mike and Giovanni,

re: logon as 'mcidas'

>I imagine you'll find something I have messed up as usual :-).

I do not expect that at all.

>One thing that
>may be of interest is that I compiled with Sun workshop 4.2 compilers, which
>I know are a bit out of date.

Interesting, but this _should_ not have anything to do with the MAP
failure for Giovanni's NCAL images, but you never know.  Are you up to
current patch levels for the SC4.2 compilers?

>Much thanks!

No worries.

Well, I found out what was going on, but not why it is going on.  What
is happening is that the full navigation block from the image is not
getting written into the frame directory on your system.  Why this is
happening I can not say yet.  What I did to find this was:

o load the 23:30 NCAL image into frame 1 on rossby:

  <as met171>
  cd mcidas/data
  mcenv -f 25@480x640
  imgdisp.k MYNCAL/AREA.25

o list out the contents of the frame directory:

  lwu.k LIST Frame1.0 0 767

o do the exact same load and list on a freshly rebuilt 7.8 distribution
  here at the UPC (on a Sun-Blade-1000 running Solaris 8 and using WS6
  C and Fortran compilers)

o do a side-by-side comparison of the frame directory contents from you
  and my machine

What I found was that the full navigation block was not being written
to the frame directory on rossby, and it is on our machine.

So, I copied over the working copy of agetserv from here to rossby:

<as 'mcidas'>
cd mcidas7.8/src
scp address@hidden:mcidas7.8/src/agetserv agetserv.upc

and made it the one used on rossby:

rm ~/bin/agetserv
ln agetserv.upc ~/bin/agetserv

Now, when I load NCAL images as user 'met171', I CAN draw maps on them!

So, the problem _must_ be somehow related to your Sun SC4.2 compilers!
This situation seems oddly familiar.  Mike, didn't we run into
something like this (an SC4.2 compilation problem) previously?

Exactly which routine(s) is(are) being compiled incorrectly is unknown
(agetserv calls a lot of subroutines to get its work done).  I will
look into this more as time permits.

Giovanni,

You now need to see if you can do the things you need to do
with the images from the NPS.  Please let me know the results.

Tom

>From address@hidden Thu Nov  8 15:53:02 2001
>To: Unidata Support <address@hidden>
>cc: Mike Voss <address@hidden>, address@hidden
>Subject: Re: 20011108: IMGDISP problems - Hallelujah!! 

YES! YES! YES! YES! YES! 

it works, when I probe the images I get the lat lon, and when I convert to
netcdf everything looks good!! ;)))

thank you so much guys!! I know I'm repeating myself, but I really
approciated your work!! ;)))

take care!!

*****************************************************************************
Giovanni Leoncini
Graduate Student
Meteorology Department                     
San Jose State University
One Washington Square,                         tel :  408-924-5199
San Jose, CA, 95112                           email:  address@hidden
*****************************************************************************